[machine translation]

machine translation, machine translations

[natural language]

natural language, natural languages

[statistical machine translation]

Statistical machine translation, statistical machine translation

[machine translation system]

machine translation systems, machine translation system

[paper]

paper, papers

[language]

language, Language, languages

[experimental results]

Experimental results, experimental results

[speech recognition]

speech recognition

[word]

words, word

[parallel corpus]

parallel corpus, parallel corpora

[method]

methods, method

[system]

systems, system

[training data]

training data

[language processing]

language processing, Language Processing

[natural language interface]

natural language interface, natural language interfaces

[approach]

approach, approaches

[statistical machine]

statistical machine

[word sense]

word sense

[MT system]

MT systems, MT system

[translation]

translation, translations, Translations

[user modeling]

user modeling

[model]

model, models

[sentence]

sentences, sentence, Sentence

[dialogue system]

dialogue systems, dialogue system

[feature structures]

feature structures

[machine]

machine

[unknown word]

unknown words, unknown word

[language model]

language model, language models

[language understanding]

language understanding, Language understanding

[user model]

user model, User models, user models

[bilingual corpus]

bilingual corpus, bilingual corpora

[result]

results, resulting, Results, result

[different inference types]

different inference types

[natural language system]

natural language system

[machine translation output]

machine translation output, machine translation outputs

[natural language processing]

natural language processing

[results of the experiments]

results of the experiment, results of the experiments

[set of features]

set of features, sets of features

[statistical machine translation system]

statistical machine translation system

[machine translation MT]

machine translation MT

[SMT model]

SMT model, SMT models

[Significant improvement]

significant improvements, Significant improvement, significant improvement

[answers to questions]

answers to questions

[information]

information, Information

[corpus]

corpora, corpus

[text]

text, texts

[e .g]

e .g

[evaluation measure]

evaluation measures, evaluation measure

[language pairs]

language pairs

[user]

user, users, User

[sophisticated representation]

sophisticated representation, sophisticated representations

[sentence planner]

sentence planner, sentence planners

[homophone errors]

homophone errors

[performance]

performance, performances, Performance

[language system]

language system, Language System

[feature]

feature, features

[grammar]

grammars, grammar, Grammars

[types of information]

types of information

[linguistically sophisticated representation]

linguistically sophisticated representation, linguistically sophisticated representations

[natural language input]

natural language input

[machine translation MT systems]

machine translation MT systems

[Statistical machine translation SMT]

Statistical machine translation SMT, statistical machine translation SMT

[100 ,000 words]

100 ,000 words

[cue phrases referring expressions]

cue phrases referring expressions

[trainable sentence planner]

trainable sentence planner

[new event detection]

new event detection

[correct syntactic structure]

correct syntactic structure

[pronoun resolution in dialogue]

pronoun resolution in dialogue

[expert human translation]

expert human translations, expert human translation

[word segmentation problem]

word segmentation problem

[speech]

speech, Speech

[error correction]

error correction, Error correction

[inference types]

inference types

[word vectors]

word vectors

[attentional state]

attentional state

[previous work]

previous work, Previous work

[translation model]

translation model, translation models

[user utterance]

user utterances, user utterance

[polynomial time]

polynomial time

[parsing strategies]

parsing strategies

[training speaker]

training speakers, training speaker

[target speaker]

target speaker

[information retrieval]

information retrieval

[two formalisms]

two formalisms

[computational complexity]

computational complexity

[Machine Translation]

Machine Translation

[evaluation]

evaluations, evaluation

[problem]

problem, problems

[experiment]

experiment, experiments, Experiments, Experiment

[understanding result]

understanding result, understanding results

[CSR corpus]

CSR corpus

[grammar formalism]

grammar formalism, grammar formalisms

[small corpus]

small corpus

[algorithm]

algorithm, algorithms

[training]

training

[discourse]

discourse, Discourse

[use]

use

[correct candidate]

correct candidates, correct candidate

[important role]

important role, important roles

[MT]

MT

[error]

errors, error, Error

[parsing]

parsing, Parsing

[dialogue]

dialogue, Dialogue

[task]

task, tasks

[subtopic boundaries]

subtopic boundaries

[Node-based inference]

Node-based inference, node-based inference

[human language]

human language

[large corpus]

large corpus

[novel method]

novel method

[linguistic structure]

linguistic structure

[parse tree]

parse tree, parse trees

[English word]

English word, English words

[answering agents]

answering agents

[human judgment]

human judgment, human judgments

[control structure]

control structure

[word string]

word strings, word string

[target language]

Target languages, target language

[Syntax-based statistical machine translation MT]

Syntax-based statistical machine translation MT, syntax-based statistical machine translation MT

[Isomorphic Grammars]

Isomorphic Grammars

[noun phrase]

noun phrases, noun phrase

[one sense]

one sense

[holding multiple candidates for understanding results]

holding multiple candidates for understanding results

[empirical results]

empirical results

[WSD models]

WSD models

[single word]

single words, single word

[correct one]

correct ones, correct one

[input documents]

input documents

[Path-based inference]

Path-based inference

[intentional structure]

intentional structure

[text understanding]

text understanding

[HDAG Kernel]

HDAG Kernel

[lexical entries]

lexical entries

[learning phrases]

learning phrases, Learning phrases

[Oral communication]

Oral communication, oral communication

[dialog systems]

dialog systems

[error-correction rules]

error-correction rules

[sense-tagged data]

sense-tagged data

[disjunctive values]

disjunctive values

[task-oriented dialogs]

task-oriented dialogs

[such phrases]

such phrases

[two tasks]

two tasks

[language translation]

language translation

[semantic constraints]

semantic constraints

[translation quality]

translation quality

[spontaneous speech]

spontaneous speech

[missing fragments]

missing fragments

[according to]

according to

[equivalent RCG]

equivalent RCGs, equivalent RCG

[Japanese homophone errors in compound nouns]

Japanese homophone errors in compound nouns

[100 words]

100 words

[lexical substitutions]

lexical substitutions

[Combination methods]

Combination methods, combination methods

[summarization quality]

summarization quality

[actor paradigm]

actor paradigm

[taxonomic similarity]

taxonomic similarity

[important contents]

important contents

[exponential time]

exponential time

[error rate]

error rate

[statistical models]

statistical models

[three strategies]

three strategies

[new domains]

new domains

[associative similarity]

associative similarity

[sentence plan]

sentence plans, sentence plan

[depending on]

depending on

[semantic equivalence]

semantic equivalence

[bilingual texts]

bilingual texts

[word alignment]

word alignment, word alignments

[translation process]

translation process

[defeasible reasoning]

defeasible reasoning

[competitive results]

competitive results

[speaker's intention]

speaker's intention

[translation system]

translation systems, translation system

[synchronous TAGs]

synchronous TAGs

[inference rules]

inference rules

[learning]

learning, Learning

[analysis]

analysis

[data]

data

[application]

application, applications, Applications

[set]

set, sets

[question]

questions, question, Question, questioning

[domain]

domain, domains

[structure]

structure, structures, Structure

[theory]

theories, theory, Theory

[phrase]

phrase, phrases

[answer]

answers, answering, answer

[document]

document, documents, Document

[work]

work, working, Working

[ambiguity]

ambiguity, ambiguities

[research]

research

[paraphrase]

paraphrase, paraphrases

[SMT]

SMT

[bigram language model]

bigram language model

[unsupervised WSD systems]

unsupervised WSD systems

[words contributing matches]

words contributing matches

[using rules in addition]

using rules in addition

[use of features]

use of features

[application of TAGs]

application of TAGs

[Thai-English language identification]

Thai-English language identification

[analysis of context languages]

analysis of context languages

[English -speaking analyst]

English -speaking analyst

[Web-based computer-assisted academic writing]

Web-based computer-assisted academic writing

[output of recognition]

output of recognition

[lesser extent entailment]

lesser extent entailment

[Penn Treebank WSJ]

Penn Treebank WSJ

[eliciting subjective human judgments]

eliciting subjective human judgments

[word-based trigram language model]

word-based trigram language model

[using various tagging strategies]

using various tagging strategies

[word sense disambiguation WSD]

word sense disambiguation WSD

[sum of the number]

sum of the number

[many more questions]

many more questions

[specific move in light]

specific move in light

[appropriate pints in MT]

appropriate pints in MT

[Automatic evaluation metrics]

Automatic evaluation metrics

[range concatenation grammar]

range concatenation grammar

[natural language question]

natural language question

[Syntactic Role System]

Syntactic Role System

[evidence of usefulness]

evidence of usefulness

[confines of syntax]

confines of syntax

[explication of inheritance]

explication of inheritance

[measure of progress]

measure of progress

[Finding the correct candidates]

Finding the correct candidates

[object-oriented grammar model]

object-oriented grammar model

[utterances of the discourse]

utterances of the discourse

[compound noun component]

compound noun component

[purposes of correction]

purposes of correction

[activities in meeting situation]

activities in meeting situation

[performance using ASR output]

performance using ASR output

[subset only 14 .0%]

subset only 14 .0%

[speaker's gaze information]

speaker's gaze information

[terminal substring parsing]

terminal substring parsing

[user's informational goals]

user's informational goals

[more linguistically sophisticated representation]

more linguistically sophisticated representation, more linguistically sophisticated representations

[factor of 5 speed-up]

factor of 5 speed-up

[two sentence planners]

two sentence planners

[basis of translation examples]

basis of translation examples

[piece of work]

piece of work

[name identification capability]

name identification capability

[English sentence structure]

English sentence structure

[challenge for the application]

challenge for the application

[reduction in error]

reduction in error

[polysemous words like sentence]

polysemous words like sentence

[least 99% correct sentences]

least 99% correct sentences

[impact of precision]

impact of precision

[foundation exploring English-Chinese bitexts]

foundation exploring English-Chinese bitexts

[probabilistic spectral mapping]

probabilistic spectral mapping

[integration of image]

integration of image

[benefits of system combination]

benefits of system combination

[exploiting linguistically sophisticated representations]

exploiting linguistically sophisticated representations

[voice interactive system]

voice interactive system

[access to text collections]

access to text collections

[complexity of documents]

complexity of documents

[resulting NE system]

resulting NE system

[needs of a publication]

needs of a publication

[rating on average]

rating on average

[only scarce resources]

only scarce resources

[choice of syntactic structure]

choice of syntactic structure

[calculus of equivalences]

calculus of equivalences

[using this novel approach]

using this novel approach

[state of the art]

state of the art

[he/she/man/woman for PERSON NE]

he/she/man/woman for PERSON NE

[instance of a pattern]

instance of a pattern

[domain of sentence condensation]

domain of sentence condensation

[method using expectations]

method using expectations

[view of language definition]

view of language definition

[Natural Language Processing]

Natural Language Processing

[LNR research group]

LNR research group

[predicting top-level subtopic boundaries]

predicting top-level subtopic boundaries

[directly from weather data]

directly from weather data

[multi-site data collection paradigm]

multi-site data collection paradigm

[natural language generation system]

natural language generation system

[user-defined performance thresholds]

user-defined performance thresholds

[CCLINC Korean-to-English translation system]

CCLINC Korean-to-English translation system

[series of experiments]

series of experiments

[source-language dependency parser]

source-language dependency parser

[manually sense-tagged data]

manually sense-tagged data

[machine translation task]

machine translation task

[chunk of information]

chunk of information

[block selection criteria]

block selection criteria

[backbone of the annotation]

backbone of the annotation

[research in generation]

research in generation

[quality of the data]

quality of the data

[concept for the NE]

concept for the NE

[basis of an ontology]

basis of an ontology

[February '92 test sentences]

February '92 test sentences

[larger training set]

larger training set

[cooccurrence of categories]

cooccurrence of categories

[large class of words]

large class of words

[two distinct tasks]

two distinct tasks

[Air Travel Information System]

Air Travel Information System

[database query applications]

database query applications

[parse reduction operating]

parse reduction operating

[responding to the user]

responding to the user

[construction of format]

construction of format

[model TDMT on APs]

model TDMT on APs

[adopting statistical techniques]

adopting statistical techniques

[large amount of speech]

large amount of speech

[use of test-score semantics]

use of test-score semantics

[redundancy elimination problem]

redundancy elimination problem

[unsupervised structure finding methods]

unsupervised structure finding methods

[noisy channel framework]

noisy channel framework

[variety of applications]

variety of applications

[non-literal aspects of communication]

non-literal aspects of communication

[high dimensionality space]

high dimensionality space

[standard statistical models]

standard statistical models

[namely the forward-backward algorithm]

namely the forward-backward algorithm

[progress in the field]

progress in the field

[monolingual parallel corpora]

monolingual parallel corpora

[linguistic generality available]

linguistic generality available

[grammar coverage problems]

grammar coverage problems

[probabilistic translation models]

probabilistic translation models

[decision tree learning]

decision tree learning

[very strong discourse effect]

very strong discourse effect

[wide coverage English grammar]

wide coverage English grammar

[discussion comparing the expressiveness]

discussion comparing the expressiveness

[abstracts of research articles]

abstracts of research articles

[appropriate user modeling]

appropriate user modeling

[framework assessing developments]

framework assessing developments

[context of use]

context of use

[range concatenation languages]

range concatenation languages

[human plausible reasoning]

human plausible reasoning

[arbiter-based combination strategies]

arbiter-based combination strategies

[knowledge of the domain]

knowledge of the domain

[most existing approaches]

most existing approaches

[resultant bilingual corpus]

resultant bilingual corpus

[problem with determiners]

problem with determiners

[user in dialogue systems]

user in dialogue systems

[task placing commas]

task placing commas

[experts in text mining]

experts in text mining

[meaning of a proposition]

meaning of a proposition

[token classification task]

token classification task

[own lexical hierarchy]

own lexical hierarchy

[focusing on noun phrases]

focusing on noun phrases

[most text processing software]

most text processing software

[baseline of 54 .55%)]

baseline of 54 .55%)

[much more available resource]

much more available resource

[result suppressing the quantifier]

result suppressing the quantifier

[approach to semantics]

approach to semantics

[intelligibility of MT output]

intelligibility of MT output

[relevance to default reasoning]

relevance to default reasoning

[set of parsing flexibilities]

set of parsing flexibilities

[such as English]

such as English

[source interval projections]

source interval projections

[corpus of sentences]

corpus of sentences

[continuous Markov models CDHMM]

continuous Markov models CDHMM

[objects of the structures]

objects of the structures

[machine translation (MT) evaluation]

machine translation (MT) evaluation

[multiple machine translation systems]

multiple machine translation systems

[overview of the research]

overview of the research

[sentence-aligned parallel corpus]

sentence-aligned parallel corpus

[learning algorithm from data]

learning algorithm from data

[procedural dialog model]

procedural dialog model

[interaction with the user]

interaction with the user

[evaluation of language learners]

evaluation of language learners

[framework of Isomorphic Grammars]

framework of Isomorphic Grammars

[principle-and-parameters language framework]

principle-and-parameters language framework

[implicit reference resolution]

implicit reference resolution

[true text through transformation]

true text through transformation

[multi-level answer resolution algorithm]

multi-level answer resolution algorithm

[question answering session]

question answering session

[phone n-gram model]

phone n-gram model

[people's answers to questions]

people's answers to questions

[resolving the ambiguity]

resolving the ambiguity

[good-quality MT system]

good-quality MT system

[issue of system response]

issue of system response

[application back end]

application back end

[series of modifications]

series of modifications

[role of user modeling]

role of user modeling

[message passing protocols]

message passing protocols

[predicting top-level boundaries]

predicting top-level boundaries

[logical form language]

logical form language

[natural language generator]

natural language generator

[method for utterance classification]

method for utterance classification

[strength of potential antecedence]

strength of potential antecedence

[bi-directional English-Spanish fragment]

bi-directional English-Spanish fragment

[NE e .g he/she/man/woman]

NE e .g he/she/man/woman

[method on a coverage]

method on a coverage

[rule-based translation systems]

rule-based translation systems

[easy word classification]

easy word classification

[information about meeting context]

information about meeting context

[unsupervised learning approach]

unsupervised learning approach

[means for lexicon specification]

means for lexicon specification

[typical kinds of users]

typical kinds of users

[additional source of constraint]

additional source of constraint

[heuristically-produced complete sentences]

heuristically-produced complete sentences

[context of a dialogue]

context of a dialogue

[polarization of the objects]

polarization of the objects

[novel customizable IE paradigm]

novel customizable IE paradigm

[intelligent interactive systems]

intelligent interactive systems

[impediment to progress]

impediment to progress

[right of the islands]

right of the islands

[novel training method]

novel training method

[so many possible structures]

so many possible structures

[power of NLP techniques]

power of NLP techniques

[most likely answer candidates]

most likely answer candidates

[such as BLEU]

such as BLEU

[other clustering methods]

other clustering methods

[behaviour of cue phrases]

behaviour of cue phrases

[translation quality of EBMT]

translation quality of EBMT

[two core modules]

two core modules

[variant of TAGs]

variant of TAGs

[general user modeling systems]

general user modeling systems

[improving OCR accuracy]

improving OCR accuracy

[very noisy portion]

very noisy portion

[proprietary stemmer above]

proprietary stemmer above

[hand-crafted template-based approaches]

hand-crafted template-based approaches

[few common noun]

few common noun

[even in the presence]

even in the presence

[SMT research community]

SMT research community

[mostly young women]

mostly young women

[reason for that difference]

reason for that difference

[determining the conversational vehicle]

determining the conversational vehicle

[Owing to the problem]

Owing to the problem

[degree of ambiguity]

degree of ambiguity

[method for discourse segmentation]

method for discourse segmentation

[using maximum likelihood method]

using maximum likelihood method

[multiple consecutive words]

multiple consecutive words

[approximations to all]

approximations to all

[twenty Switchboard dialogues]

twenty Switchboard dialogues

[block identities e .g]

block identities e .g

[NIST evaluation metric]

NIST evaluation metric

[powerful personal computers]

powerful personal computers

[retrieving examples most similar]

retrieving examples most similar

[dictionaries of word forms]

dictionaries of word forms

[Statistical Learning Model SLM]

Statistical Learning Model SLM

[modifying the transition probabilities]

modifying the transition probabilities

[large corpus resource]

large corpus resource

[algorithm on a corpus]

algorithm on a corpus

[basics of SMT]

basics of SMT

[oracle using a network]

oracle using a network

[novel classification method]

novel classification method

[document preference sets( DoPS]

document preference sets( DoPS

[frequency of events]

frequency of events

[leaving these counterexamples unexplained]

leaving these counterexamples unexplained

[specification of character size]

specification of character size

[breakdown of the performance]

breakdown of the performance

[series of tests]

series of tests

[arbitrary set of features]

arbitrary set of features

[further grammar tailoring]

further grammar tailoring

[aspects of communication]

aspects of communication

[quality of utterances]

quality of utterances

[establishing dependency relations]

establishing dependency relations

[using this approach]

using this approach, Using this approach

[attention of the participants]

attention of the participants

[issues with regard]

issues with regard

[understanding of scruffy texts]

understanding of scruffy texts

[rule-based machine translation modules]

rule-based machine translation modules

[paraphrases in one language]

paraphrases in one language

[retrieval of long phrases]

retrieval of long phrases

[standard 360K floppy]

standard 360K floppy

[dialogue translation system]

dialogue translation system

[method of structure-sharing]

method of structure-sharing

[Node-based inference rules]

Node-based inference rules

[definitions of the structure]

definitions of the structure

[use of NLP techniques]

use of NLP techniques

[number of years]

number of years

[new language pairs]

new language pairs

[current decoder implementations]

current decoder implementations

[machine learning experiments centering]

machine learning experiments centering

[discourse understanding process]

discourse understanding process

[neutral meaning representation]

neutral meaning representation

[Lexical-Functional Grammars LFG]

Lexical-Functional Grammars LFG

[sets of graphs]

sets of graphs

[U .S military]

U .S military

[most difficult SENSEVAL-2 nouns]

most difficult SENSEVAL-2 nouns

[Topic Detection tasks]

Topic Detection tasks

[middle of the scale]

middle of the scale

[using length-based criterion]

using length-based criterion

[existence of another path]

existence of another path

[leveraging human language technology]

leveraging human language technology

[sentence alignment tasks]

sentence alignment tasks

[structure of the sequence]

structure of the sequence

[maximization of translation accuracy]

maximization of translation accuracy

[set of candidate parses]

set of candidate parses

[focus post-processing the output]

focus post-processing the output

[existence of an instance]

existence of an instance

[performance of the stemmer]

performance of the stemmer

[vector space model]

vector space model

[tree adjoining grammar]

tree adjoining grammar

[modeling language acquisition]

modeling language acquisition

[interactive environment TGE]

interactive environment TGE

[order-sensitive string comparison methods]

order-sensitive string comparison methods

[using an DCG formalism]

using an DCG formalism

[Prolog a programming language]

Prolog a programming language

[example a disposition]

example a disposition

[design for conversational agents]

design for conversational agents

[strong impact on performance]

strong impact on performance

[2 a real-time response]

2 a real-time response

[examples paraphrasing ambiguous sentences]

examples paraphrasing ambiguous sentences

[average retrieval times]

average retrieval times

[natural language environment]

natural language environment

[preliminary qualitative user evaluation]

preliminary qualitative user evaluation

[relationships among intentions]

relationships among intentions

[such complex linguistic databases]

such complex linguistic databases

[many computational systems]

many computational systems

[determination of the properties]

determination of the properties

[linguistically sophisticated units]

linguistically sophisticated units

[previous papers [Zernik87]]

previous papers [Zernik87]

[experimental computer program SPRINT]

experimental computer program SPRINT

[pattern of node structures]

pattern of node structures

[problem identifying subtopic boundaries]

problem identifying subtopic boundaries

[forcing a particular interpretation]

forcing a particular interpretation

[answers to such questions]

answers to such questions

[comparable non-parallel corpora]

comparable non-parallel corpora

[English-Japanese MT system]

English-Japanese MT system

[deterministic finite automaton]

deterministic finite automaton

[simultaneously using less memory]

simultaneously using less memory

[speed of processing]

speed of processing

[logic-based grammar formalism]

logic-based grammar formalism

[decision making process]

decision making process

[sentence condensation systems]

sentence condensation systems

[predicting subtopic boundaries]

predicting subtopic boundaries

[proper treatment of syntax]

proper treatment of syntax

[translation memory system]

translation memory system

[purpose of this research]

purpose of this research

[language processing systems]

language processing systems, Language Processing systems

[37% reduction in error]

37% reduction in error

[natural language generation community]

natural language generation community

[Paramax language understanding system]

Paramax language understanding system

[numerous user studies]

numerous user studies

[ideas of Pereira]

ideas of Pereira

[cooccurrence-based word vectors]

cooccurrence-based word vectors

[less precise temporal adverbs]

less precise temporal adverbs

[strengths of each]

strengths of each

[formulas with disjunctive values]

formulas with disjunctive values

[story link detection]

story link detection

[distribution of nonverbal behaviors]

distribution of nonverbal behaviors

[Japanese homophone errors]

Japanese homophone errors

[answer resolution algorithm]

answer resolution algorithm

[number of word forms]

number of word forms

[probabilistic LR parser]

probabilistic LR parser

[more robust---an scheduling parser]

more robust---an scheduling parser

[additional increase of correlation]

additional increase of correlation

[MAP estimation approach]

MAP estimation approach

[use of lexical substitutions]

use of lexical substitutions

[human annotation performance]

human annotation performance

[grammar model for concurrent]

grammar model for concurrent

[two rule-based sentence planners]

two rule-based sentence planners

[combination with a range]

combination with a range

[recovery technique for parses]

recovery technique for parses

[implicit fuzzy quantifiers]

implicit fuzzy quantifiers

[MT evaluation techniques]

MT evaluation techniques

[predictions of missing fragments]

predictions of missing fragments

[planning algorithm into Prolog]

planning algorithm into Prolog

[modification of the document]

modification of the document

[NTHU's statistic-based system]

NTHU's statistic-based system

[natural language presuppositions]

natural language presuppositions

[dependency network representation]

dependency network representation

[human language learning process]

human language learning process

[style and/or topic]

style and/or topic

[Japanese patent claim sentences]

Japanese patent claim sentences

[evaluation of dialogue systems]

evaluation of dialogue systems

[amount of speech]

amount of speech

[conversational context features]

conversational context features

[negative impact on models]

negative impact on models

[Gaussian mixture observation densities]

Gaussian mixture observation densities

[resulting logical expression]

resulting logical expression

[text from the web]

text from the web

[application of synchronous TAGs]

application of synchronous TAGs

[many attractive properties]

many attractive properties

[arbitrarily long phrases]

arbitrarily long phrases

[much recent work]

much recent work

[existence of a number]

existence of a number

[focus on practical considerations]

focus on practical considerations

[Using this visualization method]

Using this visualization method

[monolingual material Roget's Thesaurus]

monolingual material Roget's Thesaurus

[most promising features]

most promising features

[mobile intelligent agent]

mobile intelligent agent

[laws of Hong Kong]

laws of Hong Kong

[automatic scheme collecting statistics]

automatic scheme collecting statistics

[using the parse record]

using the parse record

[Japanese revision support systems]

Japanese revision support systems

[maximum likelihood criterion]

maximum likelihood criterion

[bidirectional grammar generation system]

bidirectional grammar generation system

[frequent omissions of arguments]

frequent omissions of arguments

[variety of SMT algorithms]

variety of SMT algorithms

[principles of the system]

principles of the system

[parsing-based NE rules]

parsing-based NE rules

[such complex databases]

such complex databases

[legal text hierarchy]

legal text hierarchy

[maximum entropy classifier]

maximum entropy classifier

[76 million word]

76 million word

[Syntax-based machine translation]

Syntax-based machine translation, syntax-based machine translation

[adjusting the parameters]

adjusting the parameters

[reduction of search space]

reduction of search space

[highest levels of performance]

highest levels of performance

[attending this workshop]

attending this workshop

[stochastic tree-to-tree transducer]

stochastic tree-to-tree transducer

[terms of semantic coherence]

terms of semantic coherence

[Experiments evaluating the effectiveness]

Experiments evaluating the effectiveness

[language learning model]

language learning model

[research in user modeling]

research in user modeling

[calibrating the ranks]

calibrating the ranks

[description of implementation]

description of implementation

[literature on machine translation]

literature on machine translation

[TAP-XL Automated Analyst's Assistant]

TAP-XL Automated Analyst's Assistant

[prior to training]

prior to training

[current typical SMT models]

current typical SMT models

[statistical machine translation (SMT)]

statistical machine translation (SMT)

[synchronous dependency insertion grammar]

synchronous dependency insertion grammar, Synchronous dependency insertion grammars

[different dialog schemata]

different dialog schemata

[current statistical MT architectures]

current statistical MT architectures

[speech data from speakers]

speech data from speakers

[less than 100 words]

less than 100 words

[problem of training data]

problem of training data

[using sentence co-occurrences]

using sentence co-occurrences

[distribution of demonstrative forms]

distribution of demonstrative forms

[apartment a discourse task]

apartment a discourse task

[multimedia articulation of answers]

multimedia articulation of answers

[pronouns with NP-]

pronouns with NP-

[using organizational strategies]

using organizational strategies

[conditional loglinear models]

conditional loglinear models

[certain layout structures]

certain layout structures

[standard parser evaluation methods]

standard parser evaluation methods

[sense-tagged training data]

sense-tagged training data

[natural language data]

natural language data

[log-linear block bigram model]

log-linear block bigram model

[abstracts for digital learning]

abstracts for digital learning

[derivation forest output]

derivation forest output

[black-box OCR systems]

black-box OCR systems

[Thai key prediction]

Thai key prediction

[status of a request]

status of a request

[potential classifying SCF data]

potential classifying SCF data

[aggregation of utterances]

aggregation of utterances

[speaker group modeling]

speaker group modeling

[terms of the ability]

terms of the ability

[chemical biological warfare]

chemical biological warfare

[utilizing the numbering system]

utilizing the numbering system

[robust parsing mechanism]

robust parsing mechanism

[vocabulary size by extrapolation]

vocabulary size by extrapolation

[fair copy of texts]

fair copy of texts

[use of CCRs]

use of CCRs

[new translation model]

new translation model

[basis of a theory]

basis of a theory

[word N-gram models]

word N-gram models

[same two nodes]

same two nodes

[performance vs sentences]

performance vs sentences

[example speech recognition]

example speech recognition

[quality with paraphrases]

quality with paraphrases

[error correction methodology]

error correction methodology

[implicit intention component]

implicit intention component

[construction of domain-independent lexica]

construction of domain-independent lexica

[allowing a speaker leeway]

allowing a speaker leeway

[increasingly common speculative claim]

increasingly common speculative claim

[title-driven name recognition]

title-driven name recognition

[similar phrase-based models]

similar phrase-based models

[DARPA Communicator program]

DARPA Communicator program

[scalability against vocabulary size]

scalability against vocabulary size

[insufficient training data]

insufficient training data

[correspondences between languages]

correspondences between languages

[semantic error rate]

semantic error rate

[more remote meteorological events]

more remote meteorological events

[primarily knowledge-based mechanisms]

primarily knowledge-based mechanisms

[number of example discourses]

number of example discourses

[Most large text-understanding systems]

Most large text-understanding systems

[large number of experiments]

large number of experiments

[Wall Street Journal]

Wall Street Journal

[concept of sublanguage]

concept of sublanguage

[first project stage]

first project stage

[Unification Categorial Grammar UCG]

Unification Categorial Grammar UCG

[recall of 98%]

recall of 98%

[non-redundant lexicon specification]

non-redundant lexicon specification

[various lexical features]

various lexical features

[automatic parse-based evaluation]

automatic parse-based evaluation

[method analyzing metaphors]

method analyzing metaphors

[Senseval series of workshops]

Senseval series of workshops

[two different language pairs]

two different language pairs

[design of this module]

design of this module

[degree of success]

degree of success

[prototype concordancer CARE]

prototype concordancer CARE

[representation of the equivalence]

representation of the equivalence

[significant improvements in performance]

significant improvements in performance

[using a different LM]

using a different LM

[syntax in machine translation]

syntax in machine translation

[various structures strings]

various structures strings

[genuinely useful software infrastructure]

genuinely useful software infrastructure

[very powerful tools]

very powerful tools

[point in a conversation]

point in a conversation

[structure of nodes]

structure of nodes

[much-studied discourse task]

much-studied discourse task

[assigning ranks to referents]

assigning ranks to referents

[value of this approach]

value of this approach

[method identifying systematic patterns]

method identifying systematic patterns

[sum of each character]

sum of each character

[CSR pilot corpus]

CSR pilot corpus

[system against the data]

system against the data

[practical translation use]

practical translation use

[mixture trigram models]

mixture trigram models

[referents of the predicate]

referents of the predicate

[other current syntax formalisms]

other current syntax formalisms

[evaluation of system]

evaluation of system

[part of speech information]

part of speech information

[optical character recognition]

optical character recognition

[new similarity measures]

new similarity measures

[handling unknown words]

handling unknown words

[two distinct phases]

two distinct phases

[motivation for this effort]

motivation for this effort

[set of problems]

set of problems

[use of this approach]

use of this approach

[using a variant]

using a variant

[Web-based computer-assisted writing]

Web-based computer-assisted writing

[area of anaphora resolution]

area of anaphora resolution

[additional transfer mappings]

additional transfer mappings

[lexical information for use]

lexical information for use

[human language learners]

human language learners

[suitable superset of L]

suitable superset of L

[acquiring the knowledge]

acquiring the knowledge

[using a 5K vocabulary]

using a 5K vocabulary

[standard chart-based algorithms]

standard chart-based algorithms

[b perspective-taking in reference]

b perspective-taking in reference

[text retrieval application]

text retrieval application

[hand-crafted template-based generation component]

hand-crafted template-based generation component

[set of up extracts]

set of up extracts

[context free languages]

context free languages

[transcription errors inevitable]

transcription errors inevitable

[understanding system non-monotonic reasoning]

understanding system non-monotonic reasoning

[connection to Montague semantics]

connection to Montague semantics

[two distinct datasets]

two distinct datasets

[kinds of signals]

kinds of signals

[integration of image understanding]

integration of image understanding

[process transforming a disposition]

process transforming a disposition

[fewer mutually equivalent readings]

fewer mutually equivalent readings

[current MT architectures]

current MT architectures

[millions of features]

millions of features

[technical abstracting industry]

technical abstracting industry

[decisions on the basis]

decisions on the basis

[loss in translation quality]

loss in translation quality

[source dependency parse]

source dependency parse

[local segment contiguity models]

local segment contiguity models

[models of verbal interaction]

models of verbal interaction

[covering that lexical gap]

covering that lexical gap

[statistical machine translation quality]

statistical machine translation quality

[two parsing algorithms]

two parsing algorithms

[enrichment of human-machine interactions]

enrichment of human-machine interactions

[speech recognition hypotheses]

speech recognition hypotheses

[descriptors from individual phrases]

descriptors from individual phrases

[model of the world]

model of the world

[Arabic word segmenter]

Arabic word segmenter

[analysis of corpus]

analysis of corpus

[translation output sufficient]

translation output sufficient

[precision of 70%]

precision of 70%

[commercial applications of some]

commercial applications of some

[undisambiguated corpus data]

undisambiguated corpus data

[sole training resources]

sole training resources

[development of a model]

development of a model

[sentences in a abstract]

sentences in a abstract

[demonstration at ACL]

demonstration at ACL

[N-Best sentence hypotheses]

N-Best sentence hypotheses

[significantly better translation quality]

significantly better translation quality

[incremental grammar development]

incremental grammar development

[blocks pairs of phrases]

blocks pairs of phrases

[interesting information piece]

interesting information piece

[counterexamples to the theory]

counterexamples to the theory

[employing machine learning techniques]

employing machine learning techniques

[problems in machine translation]

problems in machine translation

[goal of this work]

goal of this work

[lexical cohesion-based approach alone]

lexical cohesion-based approach alone

[quality of the paraphrases]

quality of the paraphrases

[levels of textual representation]

levels of textual representation

[standard term weighting]

standard term weighting

[target language word segmentation]

target language word segmentation

[Traditional information retrieval techniques]

Traditional information retrieval techniques

[currently under implementation]

currently under implementation

[factor of 2 .5]

factor of 2 .5

[bigger performance gains]

bigger performance gains

[marine weather forecasts]

marine weather forecasts

[about 110 ,000 words]

about 110 ,000 words

[other linguistic tools]

other linguistic tools

[about 75% accuracy]

about 75% accuracy

[missing out words]

missing out words

[result of the comparison]

result of the comparison

[predicting new linguistic concepts]

predicting new linguistic concepts

[number of attribute sequences]

number of attribute sequences

[speech recognition word]

speech recognition word

[method interactively visualizing]

method interactively visualizing

[use in error correction]

use in error correction

[set of paraphrases]

set of paraphrases

[arbitrary HTML documents]

arbitrary HTML documents

[outputs of MT system]

outputs of MT system

[phrase boundary heuristics]

phrase boundary heuristics

[power of phrasal SMT]

power of phrasal SMT

[lexical cross-relations between English]

lexical cross-relations between English

[portable English understanding system]

portable English understanding system

[Monolingual unannotated text]

Monolingual unannotated text

[head-orientation via valency constraints]

head-orientation via valency constraints

[other Slavonic languages]

other Slavonic languages

[number of questions]

number of questions

[use of expectations]

use of expectations

[information retrieval task]

information retrieval task

[advantage of predicate-argument structures]

advantage of predicate-argument structures

[resulting spanning edges]

resulting spanning edges

[Similarity-driven Transfer System SimTran]

Similarity-driven Transfer System SimTran

[impact on performance]

impact on performance

[language learning experiment]

language learning experiment

[units of translation]

units of translation

[point of the discourse]

point of the discourse

[speaking in fragments]

speaking in fragments

[set of coherence rules]

set of coherence rules

[total processing time]

total processing time

[past 5 years]

past 5 years

[role in syntax proper]

role in syntax proper

[head-oriented notions of valency]

head-oriented notions of valency

[sum of a representation]

sum of a representation

[individual partial theories]

individual partial theories

[three relatively domain-independent capabilities]

three relatively domain-independent capabilities

[reference word string]

reference word string

[two-step clustering process]

two-step clustering process

[reappraising these insightful counterexamples]

reappraising these insightful counterexamples

[language of interest]

language of interest

[spatial descriptions in Japanese]

spatial descriptions in Japanese

[paraphrasing of multiple output]

paraphrasing of multiple output

[out-of-domain bilingual corpus]

out-of-domain bilingual corpus

[task of ambiguity resolution]

task of ambiguity resolution

[entire language system]

entire language system

[version of synchronous grammars]

version of synchronous grammars

[basis of feedback]

basis of feedback

[process describing the layout]

process describing the layout

[best other strategies]

best other strategies

[Educational Research Methods course]

Educational Research Methods course

[over 90 ,000 entries]

over 90 ,000 entries

[taking negligible runtime]

taking negligible runtime

[use of BLEU]

use of BLEU

[characteristics of this method]

characteristics of this method

[use of dependency pointers]

use of dependency pointers

[abstraction of the focus]

abstraction of the focus

[re-utterance of that portion]

re-utterance of that portion

[corpus of ,000 words]

corpus of ,000 words

[contrast to previous work]

contrast to previous work

[limited-domain computer system]

limited-domain computer system

[10 .4M English words]

10 .4M English words

[inheritance in hierarchies]

inheritance in hierarchies

[MIT Lincoln Laboratory]

MIT Lincoln Laboratory

[concurrency of computation]

concurrency of computation

[number of copies]

number of copies

[introduction of substitution costs]

introduction of substitution costs

[probabilistic context-free grammars working]

probabilistic context-free grammars working

[namely parameter smoothing]

namely parameter smoothing

[generating from such redundancy]

generating from such redundancy

[virtue of the fact]

virtue of the fact

[Statistical Machine Translation SMT]

Statistical Machine Translation SMT

[three minutes per extract]

three minutes per extract

[knowledge about category transitions]

knowledge about category transitions

[tree-based ordering model]

tree-based ordering model

[task of equivalence classification]

task of equivalence classification

[men like young women]

men like young women

[mixture of methods]

mixture of methods

[such case-based MT CBMT]

such case-based MT CBMT

[MT system design]

MT system design

[light of several aspects]

light of several aspects

[standard text browser]

standard text browser

[assessment of language pairs]

assessment of language pairs

[full linguistic analysis]

full linguistic analysis

[presence of unknown words]

presence of unknown words

[Machine Translation (MT) systems]

Machine Translation (MT) systems

[small-sized bilingual corpus]

small-sized bilingual corpus

[three-tiered approach for evaluation]

three-tiered approach for evaluation

[source language sentences]

source language sentences

[coordination using parsing strategies]

coordination using parsing strategies

[unification-based grammar formalism]

unification-based grammar formalism, Unification-based grammar formalisms

[highly inflective languages]

highly inflective languages

[syntactic ambiguity resolution]

syntactic ambiguity resolution

[inheritance as a means]

inheritance as a means

[Markov models HMM]

Markov models HMM

[IE-enhanced text browsers]

IE-enhanced text browsers

[certainty of the conclusions]

certainty of the conclusions

[structure of purposes]

structure of purposes

[new users of tool]

new users of tool

[even larger improvements]

even larger improvements

[Air Travel Information Service]

Air Travel Information Service

[second language learning]

second language learning

[knowledge-based acquisition of grammars]

knowledge-based acquisition of grammars

[two evaluation measures]

two evaluation measures

[Japanese kanji-kana characters]

Japanese kanji-kana characters

[utterance classification performance]

utterance classification performance

[broad coverage lexicon]

broad coverage lexicon

[individual lexical rules]

individual lexical rules

[more effective CFG filter]

more effective CFG filter

[semantics for feature structures]

semantics for feature structures

[word sense disambigation models]

word sense disambigation models

[adaptive dynamic word formation]

adaptive dynamic word formation

[object-oriented natural language parsing]

object-oriented natural language parsing

[kind of word vectors]

kind of word vectors

[different information in memory]

different information in memory

[current SMT models]

current SMT models

[multilingual multimedia data]

multilingual multimedia data

[machine learning approach]

machine learning approach

[starting point the beginning]

starting point the beginning

[issue of domain dependence]

issue of domain dependence

[precision of 96%]

precision of 96%

[database query paraphrase]

database query paraphrase

[crucial issue recently]

crucial issue recently

[advantage of the sparsity]

advantage of the sparsity

[several levels of chunks]

several levels of chunks

[purpose building spelling-checkers]

purpose building spelling-checkers

[improving machine translation outputs]

improving machine translation outputs

[significantly better results]

significantly better results

[sentence-level semantic equivalence classification]

sentence-level semantic equivalence classification

[other approaches to WSI]

other approaches to WSI

[translations of each other]

translations of each other

[uncertainty in expert systems]

uncertainty in expert systems

[applications of this technology]

applications of this technology

[word-dependent substitution costs]

word-dependent substitution costs

[clustering polysemic verbs]

clustering polysemic verbs

[working computer program]

working computer program

[information in English]

information in English

[Using these ideas together]

Using these ideas together

[Conceptual Dependency Diagram CDD]

Conceptual Dependency Diagram CDD

[performance of an oracle]

performance of an oracle

[Various properties of discourse]

Various properties of discourse

[knowledge of surface English]

knowledge of surface English

[knowledge-based information retrieval]

knowledge-based information retrieval

[such as words]

such as words

[papers in English]

papers in English

[English-Japanese machine translation]

English-Japanese machine translation

[statistical models of WH-questions]

statistical models of WH-questions

[underlying linguistic theory]

underlying linguistic theory

[approach of KPSG]

approach of KPSG

[two stages of 1]

two stages of 1

[natural language texts]

natural language texts

[log-linear maximum-entropy models]

log-linear maximum-entropy models

[underlying concurrent computation model]

underlying concurrent computation model

[robust island-based parsing method]

robust island-based parsing method

[poor syntactic construction]

poor syntactic construction

[boundary between root]

boundary between root

[adaptive learning procedure]

adaptive learning procedure

[success of ensemble methods]

success of ensemble methods

[quality of multilingual summaries]

quality of multilingual summaries

[existing probabilistic parser]

existing probabilistic parser

[language pairs like English-Chinese]

language pairs like English-Chinese

[SmTA double check]

SmTA double check

[Path-based inference rules]

Path-based inference rules, path-based inference rules

[individual coordinate conjunction]

individual coordinate conjunction

[block unigram model]

block unigram model

[full linguistic analysis component]

full linguistic analysis component

[long distance constraints]

long distance constraints

[costly dependency pointers]

costly dependency pointers

[model analyzing English sentences]

model analyzing English sentences

[part of the parser]

part of the parser

[Current natural language interfaces]

Current natural language interfaces

[single Intel i860 chip]

single Intel i860 chip

[part of a unit]

part of a unit

[orders of magnitude]

orders of magnitude

[correcting minor errors]

correcting minor errors

[description of feature structures]

description of feature structures, descriptions of feature structures

[definite noun phrase reiteration]

definite noun phrase reiteration

[simple mixture language model]

simple mixture language model

[traditional statistical approaches]

traditional statistical approaches

[two different algorithms]

two different algorithms

[Hidden Markov Model]

Hidden Markov Model

[one unique author]

one unique author

[basics of monolingual UCG]

basics of monolingual UCG

[disjunctive normal form]

disjunctive normal form

[one of these theories]

one of these theories

[set of inter-related tasks]

set of inter-related tasks

[structure-sharing of graphs]

structure-sharing of graphs

[new CSR corpus]

new CSR corpus

[grammar checker for Basque]

grammar checker for Basque

[Case Marker System]

Case Marker System

[MIT ATIS system]

MIT ATIS system

[experiments in this article]

experiments in this article

[Horn logic program]

Horn logic program

[method for error correction]

method for error correction

[using a neural network]

using a neural network

[2 .284 SRHs]

2 .284 SRHs

[small Arabic corpus]

small Arabic corpus

[notions of noun group]

notions of noun group

[including Natural Language Processing]

including Natural Language Processing

[Offering advantages like reproducability]

Offering advantages like reproducability

[grammar for this generator]

grammar for this generator

[document understanding technology]

document understanding technology

[performance of the system]

performance of the system

[sequence of utterances]

sequence of utterances

[32 .8% improvement]

32 .8% improvement

[domain of Navy messages]

domain of Navy messages

[operation of the mechanisms]

operation of the mechanisms

[several phrase-based translation models]

several phrase-based translation models

[approximation of HPSG]

approximation of HPSG

[difference between sentence structure]

difference between sentence structure

[formalization of defeasible reasoning]

formalization of defeasible reasoning

[reasonably neat form]

reasonably neat form

[assumption in neat form]

assumption in neat form

[large unsegmented Arabic corpus]

large unsegmented Arabic corpus

[dynamic language model combination]

dynamic language model combination

[carefully controlling the order]

carefully controlling the order

[empirical conversation analysis]

empirical conversation analysis

[degree of hastiness]

degree of hastiness

[other MT systems]

other MT systems

[February '92 benchmark evaluation]

February '92 benchmark evaluation

[using concept-based seeds]

using concept-based seeds

[importance of the issue]

importance of the issue

[discourse understanding accuracy]

discourse understanding accuracy

[knowledge about task-oriented dialogs]

knowledge about task-oriented dialogs

[disjoint English phrases]

disjoint English phrases

[manual evaluation of strings]

manual evaluation of strings

[behavior of the oracle]

behavior of the oracle

[data collection effort]

data collection effort

[very simple sentence-plan-generator SPG]

very simple sentence-plan-generator SPG

[underspecified chart representations]

underspecified chart representations

[broad range of texts]

broad range of texts

[special method describing inflection]

special method describing inflection

[various relations between entities]

various relations between entities

[lack of automatic methods]

lack of automatic methods

[performance of systems]

performance of systems

[continuous speech recognition]

continuous speech recognition

[kind of similarity]

kind of similarity, kinds of similarity

[future research topics]

future research topics

[training reference speaker]

training reference speaker

[effectiveness of these indices]

effectiveness of these indices

[bilingual parallel corpus]

bilingual parallel corpora, bilingual parallel corpus

[only bilingual resource]

only bilingual resource

[word-semantic information e .g]

word-semantic information e .g

[DARPA Resource Management corpus]

DARPA Resource Management corpus

[examples of the difference]

examples of the difference

[using part-of-speech tag sequences]

using part-of-speech tag sequences

[TREC 2003 QA tracks]

TREC 2003 QA tracks

[context-dependent phonetic modelling]

context-dependent phonetic modelling

[focus of attention]

focus of attention

[hope generating the parse]

hope generating the parse

[last two decades]

last two decades

[method resolving this ambiguity]

method resolving this ambiguity

[natural language generation]

natural language generation

[theory of attention]

theory of attention

[robust parsing capability]

robust parsing capability

[Category Cooccurrence Restrictions CCRs]

Category Cooccurrence Restrictions CCRs

[support to the use]

support to the use

[disambiguating the nouns]

disambiguating the nouns

[prior to content determination]

prior to content determination

[completely missing fragments]

completely missing fragments

[standard evaluation metric]

standard evaluation metric

[CSR Speech Recognition corpus]

CSR Speech Recognition corpus

[demonstrative expressions in English]

demonstrative expressions in English

[Japanese sentence structure]

Japanese sentence structure

[distinction among these components]

distinction among these components

[use of sophisticated representations]

use of sophisticated representations

[language processing modules]

language processing modules

[recognition with this model]

recognition with this model

[young men like women]

young men like women

[consistency problem for formulas]

consistency problem for formulas

[relationships between language]

relationships between language

[instance statistical MT systems]

instance statistical MT systems

[novel approach to Q/A]

novel approach to Q/A

[top human-ranked sentence plan]

top human-ranked sentence plan

[prior RCL parser]

prior RCL parser

[context of a task]

context of a task

[introducing this approach]

introducing this approach

[useful disambiguation tool]

useful disambiguation tool

[number of comparative evaluations]

number of comparative evaluations

[Arabic's rich morphology]

Arabic's rich morphology

[baseline model’s score]

baseline model’s score

[unsupervised word alignment component]

unsupervised word alignment component

[placement of function words]

placement of function words

[error for the purposes]

error for the purposes

[natural language parsing NLP]

natural language parsing NLP

[errors in machine translation]

errors in machine translation

[management of uncertainty]

management of uncertainty

[18 .3M Chinese characters]

18 .3M Chinese characters

[method detecting speech repairs]

method detecting speech repairs

[machine learning techniques]

machine learning techniques

[longest possible span]

longest possible span

[IBM Models 1-2]

IBM Models 1-2

[standard Arabic-English translation task]

standard Arabic-English translation task

[sentence error rates]

sentence error rates

[enlarging the separation margin]

enlarging the separation margin

[different language families]

different language families

[according to individual needs]

according to individual needs

[concept of a syllogism]

concept of a syllogism

[new flexible inference method]

new flexible inference method

[arcs between two nodes]

arcs between two nodes

[sets of subpredicates]

sets of subpredicates

[looking up phrase translations]

looking up phrase translations

[target recognition task]

target recognition task

[new interactive disambiguation scheme]

new interactive disambiguation scheme

[construction of statistical models]

construction of statistical models

[accurate word order generation]

accurate word order generation

[yet more redundancy]

yet more redundancy

[polynomial time decoding algorithm]

polynomial time decoding algorithm

[Generation of Referring Expressions]

Generation of Referring Expressions

[generating the correct parse]

generating the correct parse

[use of demonstrative expressions]

use of demonstrative expressions

[avoiding failures of reference]

avoiding failures of reference

[internal lexical variation]

internal lexical variation

[submanifold of data]

submanifold of data

[various interesting ambiguities]

various interesting ambiguities

[Most importantly Chinese-English texts]

Most importantly Chinese-English texts

[connecting referents with predicates]

connecting referents with predicates

[terms of retrieval accuracy]

terms of retrieval accuracy

[high quality IE results]

high quality IE results

[corpus of words]

corpus of words

[examples of parses]

examples of parses

[input Arabic documents]

input Arabic documents

[dictionary-based word vectors]

dictionary-based word vectors

[selection of fonts]

selection of fonts

[to reasoning with dispositions]

to reasoning with dispositions

[accurate predicate-argument structures]

accurate predicate-argument structures

[potential of text browsers]

potential of text browsers

[current state of research]

current state of research

[target word selection]

target word selection

[vagueness in semantic annotation]

vagueness in semantic annotation

[Heuristic Parsing Model HPM]

Heuristic Parsing Model HPM

[performance for sophisticated representations]

performance for sophisticated representations

[Syllogistic reasoning with dispositions]

Syllogistic reasoning with dispositions

[case of multi-document summarization]

case of multi-document summarization

[verbal grounding acts]

verbal grounding acts

[sentence utilizing examples effectively]

sentence utilizing examples effectively

[explanations for the behaviour]

explanations for the behaviour

[single understanding result]

single understanding result

[using a trigram model]

using a trigram model

[path between the nodes]

path between the nodes

[treelet translation pairs]

treelet translation pairs

[movements of word blocks]

movements of word blocks

[parsimonious instantiation scheme]

parsimonious instantiation scheme

[operation of the system]

operation of the system

[new evaluation measure]

new evaluation measure

[training two successive learners]

training two successive learners

[new target speaker]

new target speaker

[computational complexity of unification]

computational complexity of unification

[using real dialogue data]

using real dialogue data

[reasonable classification accuracy]

reasonable classification accuracy

[documents over a representation]

documents over a representation

[attentional state dynamic]

attentional state dynamic

[building an SMT system]

building an SMT system

[various rhetorical functions]

various rhetorical functions

[resource sensitive logic]

resource sensitive logic

[improvement of word]

improvement of word

[October dry-run test set]

October dry-run test set

[head-to-head tests against one]

head-to-head tests against one

[last few years]

last few years

[nearest neighbour methods]

nearest neighbour methods

[solution to these problems]

solution to these problems

[histogram of keywords]

histogram of keywords

[missing words ellipsis]

missing words ellipsis

[description of an approach]

description of an approach

[Performance of the algorithm]

Performance of the algorithm

[strength of approach]

strength of approach

[method producing such phrases]

method producing such phrases

[application of this approach]

application of this approach

[design of unification algorithms]

design of unification algorithms

[working translation system]

working translation system

[machine translation pipeline]

machine translation pipeline

[complex natural language processing]

complex natural language processing

[categories in local trees]

categories in local trees

[much recent debate]

much recent debate

[using translation probabilities]

using translation probabilities

[field of speech processing]

field of speech processing

[use of lexical features]

use of lexical features

[blocks with orientation]

blocks with orientation

[accurate structural disambiguation]

accurate structural disambiguation

[DARPA speech recognition technology]

DARPA speech recognition technology

[explanation of an ambiguity]

explanation of an ambiguity

[in-domain monolingual corpus]

in-domain monolingual corpus

[using 109 training speakers]

using 109 training speakers

[LOGON MT demonstrator]

LOGON MT demonstrator

[lack of negative feedback]

lack of negative feedback

[publication of papers]

publication of papers

[terms of Dynamic Grammars]

terms of Dynamic Grammars

[operating on parse forests]

operating on parse forests

[validity of lexical entries]

validity of lexical entries

[scale of existing spelling-checkers]

scale of existing spelling-checkers

[duplicating the experiment]

duplicating the experiment

[theory of language presuppositions]

theory of language presuppositions

[fundamental problems of SMT]

fundamental problems of SMT

[novel evaluation scheme]

novel evaluation scheme

[honorifics respectful words]

honorifics respectful words

[same source text]

same source text

[relatively free word order]

relatively free word order

[However the only area]

However the only area

[framework describing the processing]

framework describing the processing

[general preference of approach]

general preference of approach

[disjunctive feature structures]

disjunctive feature structures

[using complicated large knowledgebases]

using complicated large knowledgebases

[clustering of word co-occurrences]

clustering of word co-occurrences

[word-sense disambiguation algorithm]

word-sense disambiguation algorithm

[instead of words]

instead of words

[method acquiring sense-tagged data]

method acquiring sense-tagged data

[system response to users]

system response to users

[temporal relations between segments]

temporal relations between segments

[English-Chinese parallel corpora]

English-Chinese parallel corpora

[form of N-grams]

form of N-grams

[unit on protocol analysis]

unit on protocol analysis

[certain kinds of phrases]

certain kinds of phrases

[using the Information Bottleneck]

using the Information Bottleneck

[dependency structure of sentences]

dependency structure of sentences

[Prolog subset of logic]

Prolog subset of logic

[methods achieving robustness]

methods achieving robustness

[Informally a proposition]

Informally a proposition

[method for computational analysis]

method for computational analysis

[and/or answer levels]

and/or answer levels

[CMU's SMT system]

CMU's SMT system

[list of word strings]

list of word strings

[results of the evaluations]

results of the evaluations

[predictive performance of models]

predictive performance of models

[evaluating WSD programs]

evaluating WSD programs

[using domain preference knowledge]

using domain preference knowledge

[approach to reasoning]

approach to reasoning

[hand-built symbolic resources]

hand-built symbolic resources

[segments of the discourse]

segments of the discourse

[worse coverage than accounts]

worse coverage than accounts

[processing examples in context]

processing examples in context

[local phrase re-ordering]

local phrase re-ordering

[precision on both systems]

precision on both systems

[recognizing the intentions]

recognizing the intentions

[resolving zero pronouns]

resolving zero pronouns

[non-native language essays]

non-native language essays

[natural language parsing]

natural language parsing

[utility of this constraint]

utility of this constraint

[phrases with gaps]

phrases with gaps

[shortcomings of different strategies]

shortcomings of different strategies

[subsets of larger database]

subsets of larger database

[language as a pivot]

language as a pivot

[statement of generalizations]

statement of generalizations

[statistical translation model]

statistical translation model

[segmental k-means algorithm]

segmental k-means algorithm

[especially personal names]

especially personal names

[correct chain of syllables]

correct chain of syllables

[selection of bag-of-words]

selection of bag-of-words

[idea behind method]

idea behind method

[implementation of the model]

implementation of the model

[computational complexity of some]

computational complexity of some

[conversation for documentation]

conversation for documentation

[predicate calculus notation]

predicate calculus notation

[method of abductive inference]

method of abductive inference

[almost inevitable inferences]

almost inevitable inferences

[yielding different ways]

yielding different ways

[approach to pronoun resolution]

approach to pronoun resolution

[problem extracting specific information]

problem extracting specific information

[two baseline sentence planners]

two baseline sentence planners

[balance matching operation]

balance matching operation

[Finite State Model FSM]

Finite State Model FSM

[trigram language model]

trigram language model

[valid semantic interpretation]

valid semantic interpretation

[sophisticated representation of documents]

sophisticated representation of documents, sophisticated representations of documents

[most difficult problems]

most difficult problems

[Chinese Natural Language Processing]

Chinese Natural Language Processing

[existing spelling-checkers for English]

existing spelling-checkers for English

[two kinds of similarity]

two kinds of similarity

[more sophisticated annotation]

more sophisticated annotation

[suffix array-based data structure]

suffix array-based data structure

[many classical grammatical formalisms]

many classical grammatical formalisms

[left corner parsing algorithm]

left corner parsing algorithm

[answers to semantic questions]

answers to semantic questions

[framework of international collaborations]

framework of international collaborations

[results about the effectiveness]

results about the effectiveness

[temporal anaphora resolution]

temporal anaphora resolution

[same test set]

same test set

[verbal means for grounding]

verbal means for grounding

[order]

order, orders

[novel approach]

novel approach

[multiple candidates]

multiple candidates

[anaphora resolution]

anaphora resolution

[phrase translations]

phrase translations

[referring expressions]

Referring Expressions, referring expressions

[verbal interaction]

verbal interaction, verbal interactions

[utterance]

utterance, utterances

[knowledge]

knowledge

[improvement]

improvement, improvements

[inference]

inference, inferences

[input]

input, inputs

[example]

examples, example, Examples

[target]

target, Target

[meaning]

meaning, meanings

[technique]

techniques, technique, Techniques

[parser]

parser

[dependency]

dependency

[output]

output, outputs

[time]

time, times

[speaker]

speaker's, speakers, speaker

[reference]

reference, references

[way]

way, ways

[decision]

decisions, decision

[generation]

generation, Generation

[representation]

representation, representations

[interpretation]

interpretations, interpretation

[context]

context, contexts

[concept]

concept, concepts

[tracking algorithm]

tracking algorithm

[implementation in a voice system]

implementation in a voice system

[minor extensions of pre-existing components]

minor extensions of pre-existing components

[bag-of-words methods]

bag-of-words methods

[experimental research]

experimental research

[definition of typicality a concept]

definition of typicality a concept

[Head Grammars]

Head Grammars

[m-component mixture of trigram models]

m-component mixture of trigram models

[empirical viewpoint]

empirical viewpoint

[semantic constraints in broad domains]

semantic constraints in broad domains

[suppressing the fuzzy quantifier most]

suppressing the fuzzy quantifier most

[low precision]

low precision

[complete interpretations]

complete interpretations

[Chart parsing]

Chart parsing, chart Parsing

[template-based generation]

template-based generation

[many speakers prior to training]

many speakers prior to training

[log( d]

log( d

[roughly corresponding in a chart]

roughly corresponding in a chart

[many cases]

many cases

[appropriate database]

appropriate database

[practical approach employing n-gram models]

practical approach employing n-gram models

[polarity-tagged corpus]

polarity-tagged corpus

[eye gaze]

eye gaze

[CFG filtering]

CFG filtering, CFG filter

[copies of the derivation tree]

copies of the derivation tree

[extending the n-best speech/language integration architecture]

extending the n-best speech/language integration architecture

[Naive Bayes]

Naive Bayes

[earlier ones]

earlier ones

[Multimedia answers]

Multimedia answers

[mathematical expressions]

mathematical expressions

[syntactic descriptions]

syntactic descriptions

[knowledge about the system users]

knowledge about the system users

[additional indices such as the time]

additional indices such as the time

[Experiment results]

Experiment results

[spatial concepts]

spatial concepts

[other models]

other models

[access to a pharmaceutical news archive]

access to a pharmaceutical news archive

[dependency trees]

dependency trees

[directing the process translating a sentence]

directing the process translating a sentence

[using the unification-based grammar formalism]

using the unification-based grammar formalism

[system for incremental grammar development]

system for incremental grammar development

[other documents]

other documents

[similar criteria]

similar criteria

[following ideas]

following ideas

[basic goal building that editor]

basic goal building that editor

[JAVELIN system]

JAVELIN system

[information about the language learning process]

information about the language learning process

[HK laws]

HK laws

[direct application of existing metrics]

direct application of existing metrics

[theory about these probably-correct rules]

theory about these probably-correct rules

[dialogue state]

dialogue state

[conventional algorithms for the analysis]

conventional algorithms for the analysis

[.1% a 45% reduction in error]

.1% a 45% reduction in error

[Korean-to-English machine translation system CCLINC]

Korean-to-English machine translation system CCLINC

[system's response]

system's response

[ranking tasks]

ranking tasks

[success of the parsing mechanism]

success of the parsing mechanism

[lexical weighting of phrase translations]

lexical weighting of phrase translations

[incorporating an approximation of the analysis]

incorporating an approximation of the analysis

[language definition]

language definition

[evaluating QA system performance on questions]

evaluating QA system performance on questions

[semantic interpretation]

semantic interpretation

[two nodes]

two nodes

[such sentences]

such sentences

[formalism of Category Cooccurrence Restrictions]

formalism of Category Cooccurrence Restrictions

[topical report]

topical report

[path values]

path values

[averaging the statistics of models]

averaging the statistics of models

[right translation of the words]

right translation of the words

[increasing the parsing accuracy rate]

increasing the parsing accuracy rate

[pattern prefix*-stem-suffix*]

pattern prefix*-stem-suffix*

[assuming greater]

assuming greater

[use of machine translation systems]

use of machine translation systems

[references of the pronoun in sentences]

references of the pronoun in sentences

[linguistic concepts]

linguistic concepts

[new interactive corpus exploration tool]

new interactive corpus exploration tool

[demonstrative expressions]

demonstrative expressions

[isomorphic derivations]

isomorphic derivations

[human listeners]

human listeners

[performance of a search engine]

performance of a search engine

[same discourse]

same discourse

[fluent speech]

fluent speech

[stochastic processes]

stochastic processes

[error for the language system]

error for the language system

[final constituent]

final constituent

[conjunction with structure finding methods]

conjunction with structure finding methods

[complex task]

complex task

[HTML documents]

HTML documents

[dialogue move]

dialogue move

[abstract moves]

abstract moves

[traditional word]

traditional word

[set of machine learning experiments]

set of machine learning experiments

[parallel data]

parallel data

[bootstrapping procedure]

bootstrapping procedure

[test]

test, tests

[relative effects of segment order]

relative effects of segment order

[certain sublanguages with text structure]

certain sublanguages with text structure

[true text]

true text

[Syntactic Role]

Syntactic Role

[processing accounts]

processing accounts

[non-terminal edges]

non-terminal edges

[activity discussing]

activity discussing

[multilingual summaries]

multilingual summaries

[independent event-based extractive mutli-document summarization approaches]

independent event-based extractive mutli-document summarization approaches

[surface representation of domain entities]

surface representation of domain entities

[treatment for Japanese sentence analyses]

treatment for Japanese sentence analyses

[over a third of the sentences]

over a third of the sentences

[typing location]

typing location

[effective way improving system performance]

effective way improving system performance

[prior work]

prior work

[only way]

only way

[extracting specific information from unrestricted texts]

extracting specific information from unrestricted texts

[Examination of the effect of features]

Examination of the effect of features

[such decoding]

such decoding

[multiple tokens]

multiple tokens

[according to the average precision metric]

according to the average precision metric

[principled techniques for syntactic analysis]

principled techniques for syntactic analysis

[experimental objects]

experimental objects

[then performing cluster number estimation]

then performing cluster number estimation

[member of the semantic set]

member of the semantic set

[Task-based evaluation using Arabic information retrieval]

Task-based evaluation using Arabic information retrieval

[ill-formed input]

ill-formed input

[large vocabulary]

large vocabulary

[interface browsing]

interface browsing

[using the new SI corpus]

using the new SI corpus

[casting doubt on this common assumption]

casting doubt on this common assumption

[lexical features]

lexical features

[tagging strategies]

tagging strategies

[such miscommunications]

such miscommunications

[NIST automatic MT evaluation software]

NIST automatic MT evaluation software

[free indexation]

free indexation

[average precision over unstemmed text]

average precision over unstemmed text

[Several experiments corresponding to semantic domains]

Several experiments corresponding to semantic domains

[7 .5% word error rate]

7 .5% word error rate

[new model]

new model

[English alphabets]

English alphabets

[generation modules]

generation modules

[goal naturally sounding dialog systems]

goal naturally sounding dialog systems

[Using the BU recognition system]

Using the BU recognition system

[same concepts]

same concepts

[large extent]

large extent

[important role in human cognition]

important role in human cognition

[anaphoric expressions]

anaphoric expressions

[parallel sentences in comparable non-parallel corpora]

parallel sentences in comparable non-parallel corpora

[text-plan input]

text-plan input

[unstemmed text]

unstemmed text

[unification framework]

unification framework

[automatic detection]

automatic detection

[automatic translation of natural language]

automatic translation of natural language

[recognition task]

recognition task

[most benefit from the integration]

most benefit from the integration

[provably optimal free indexation algorithm]

provably optimal free indexation algorithm

[model parameters with the vocabulary]

model parameters with the vocabulary

[transforming a disposition into a proposition]

transforming a disposition into a proposition

[grammaticality of the system output]

grammaticality of the system output

[implementing a user modeling component]

implementing a user modeling component

[dialogue duration for skilled users]

dialogue duration for skilled users

[combining several language models LMs]

combining several language models LMs

[independently valuable general-purpose NLP components]

independently valuable general-purpose NLP components

[automatic alignments]

automatic alignments

[event map]

event map

[meaning structure on the sentence]

meaning structure on the sentence

[word-based alignments]

word-based alignments

[base parser]

base parser

[dialogue corpora]

dialogue corpora

[linguistic annotation]

linguistic annotation

[knowledge level on the target domain]

knowledge level on the target domain

[entries from machine readable dictionaries]

entries from machine readable dictionaries

[model TDMT]

model TDMT

[affix removal]

affix removal

[model parameters]

model parameters

[rule-based systems]

rule-based systems

[system support of mission success]

system support of mission success

[output of black-box OCR systems]

output of black-box OCR systems

[proprietary stemmer]

proprietary stemmer

[scenario templates]

scenario templates

[two corpora]

two corpora

[scruffy texts]

scruffy texts

[high performance in real applications]

high performance in real applications

[grammar model]

grammar model

[predominant senses]

predominant senses

[correct structure]

correct structure

[multi-strategy approach]

multi-strategy approach

[other things]

other things

[phrase-based models]

phrase-based models

[using a fake non-terminal symbol]

using a fake non-terminal symbol

[digital learning]

digital learning

[test set]

test set

[simple unsupervised technique learning morphology]

simple unsupervised technique learning morphology

[Cooperative Model]

Cooperative Model

[predictive performance]

predictive performance

[promising approach]

promising approach

[human understanding of natural language]

human understanding of natural language

[diagnostic tool]

diagnostic tool

[young women]

young women

[Using alignment techniques from machine translation]

Using alignment techniques from machine translation

[experimental evaluation]

experimental evaluation, Experimental evaluation

[building translation systems for language pairs]

building translation systems for language pairs

[baseline system]

baseline system

[denotational semantics]

denotational semantics

[fundamental problem]

fundamental problems, fundamental problem

[commercial applications]

commercial applications

[parse record of the utterance]

parse record of the utterance

[computations involving]

computations involving

[MT evaluation]

MT evaluation

[Special attention]

Special attention

[use of the discourse constraint]

use of the discourse constraint

[using an underlying word alignment]

using an underlying word alignment

[alternative markers]

alternative markers

[large-scale acquisition]

large-scale acquisition

[Sentence ambiguities]

Sentence ambiguities

[larger volume]

larger volume

[missing words]

missing words

[other's proposals]

other's proposals

[linguistic theory]

linguistic theory

[initial ranking of these parses]

initial ranking of these parses

[document understanding]

document understanding

[Arabic-English translation]

Arabic-English translation

[relative performances]

relative performances

[individual users]

individual users

[alignment algorithm]

alignment algorithm

[sample output an expert translation]

sample output an expert translation

[state-of-the-art word sense disambiguation model]

state-of-the-art word sense disambiguation model

[variant of a calculus notation]

variant of a calculus notation

[disambiguation WSD]

disambiguation WSD

[information extraction]

information extraction

[using an out-of-domain bilingual corpus]

using an out-of-domain bilingual corpus

[other areas of natural language processing]

other areas of natural language processing

[forming an utterance about a task]

forming an utterance about a task

[predictive questioning]

predictive questioning

[baseline model]

baseline model

[Several experiments]

Several experiments, several experiments

[concept of a char t]

concept of a char t

[experiments showing the WSD accuracy lower]

experiments showing the WSD accuracy lower

[other resources]

other resources

[WSD models]

WSD models

[important bearing on commonsense reasoning]

important bearing on commonsense reasoning

[different reasons]

different reasons

[large number]

large number

[126 sentences]

126 sentences

[few speakers]

few speakers

[application of ambiguity packing techniques]

application of ambiguity packing techniques

[posing special problems for readers]

posing special problems for readers

[evidence scalable]

evidence scalable

[ungrammatical input]

ungrammatical input

[prediction model]

prediction model

[document representation]

document representation

[document summarization]

document summarization

[recognition network]

recognition network

[applying general-purpose language understanding technology]

applying general-purpose language understanding technology

[semantic frame]

semantic frame

[referential information]

referential information

[finding more data relevant to task]

finding more data relevant to task

[path of arcs between nodes]

path of arcs between nodes

[approximation of the formal analysis]

approximation of the formal analysis

[automatic word sense induction WSI]

automatic word sense induction WSI

[inter-related tasks]

inter-related tasks

[Multimodal Communication]

Multimodal Communication

[original question]

original question

[new approach]

new approach

[such movements]

such movements

[data collection]

data collection

[accuracy rate]

accuracy rate

[resulting algorithm]

resulting algorithm

[using class-dependent interpolation of N-grams]

using class-dependent interpolation of N-grams

[face-to-face meetings using Bayesian Network classifiers]

face-to-face meetings using Bayesian Network classifiers

[conversation transcripts]

conversation transcripts

[speech repairs]

speech repairs

[non-parallel corpus]

non-parallel corpora, non-parallel corpus

[non-constituent coordination]

non-constituent coordination

[generation system]

generation system, generation systems

[distinguishing features]

distinguishing features

[present natural language search engines]

present natural language search engines

[Human-Machine Communication including Language Processing]

Human-Machine Communication including Language Processing

[phone-string classifier]

phone-string classifier

[human text]

human text

[raw text]

raw text

[fulfillment of logical well-formedness conditions]

fulfillment of logical well-formedness conditions

[stochastic output]

stochastic output

[LRE project]

LRE project

[techniques superior]

techniques superior

[pronoun seeds]

pronoun seeds

[tool kit]

tool kit

[edges adjacent]

edges adjacent

[final structure]

final structure

[retrieval time]

retrieval times, retrieval time

[rough drafts]

rough drafts

[syntactic analysis]

syntactic analysis

[easy exploitation of limited domain semantics]

easy exploitation of limited domain semantics

[describing the layout of an apartment]

describing the layout of an apartment

[words with multiple meanings ambiguity]

words with multiple meanings ambiguity

[vital requirements of language translation]

vital requirements of language translation

[Natural Language]

Natural Language

[focusing especially on the changes]

focusing especially on the changes

[high accuracy]

high accuracy

[feature generation]

feature generation

[technical abstracting]

technical abstracting

[similarity measure]

similarity measure, similarity measures

[major new problems including vocabulary]

major new problems including vocabulary

[retrieval accuracy]

retrieval accuracy

[non-contiguous phrases]

non-contiguous phrases

[segment contiguity]

segment contiguity

[SYSCONJ system]

SYSCONJ system

[actual applications]

actual applications

[input sentence]

input sentence

[logistics system]

logistics system

[observation complete]

observation complete

[educational tool]

educational tool

[multi-document summarization]

multi-document summarization

[containing 28 ,449 word tokens]

containing 28 ,449 word tokens

[analyzing how]

analyzing how

[computational problems]

computational problems

[discourse-relevant purposes]

discourse-relevant purposes

[new paradigm for human interaction]

new paradigm for human interaction

[skilled users]

skilled users

[gathering information related to a scenario]

gathering information related to a scenario

[dependency grammars]

dependency grammars

[boosting approach]

boosting approach

[combination with conventional SMT models]

combination with conventional SMT models

[part-of-speech tagger]

part-of-speech tagger

[Experiments using the SemCor data sets]

Experiments using the SemCor data sets

[regard to the evaluation methodology]

regard to the evaluation methodology

[correlation of the new measure]

correlation of the new measure

[best way]

best way

[natural language in natural settings]

natural language in natural settings

[cohesive text]

cohesive text

[efficiency—to work]

efficiency—to work

[Korean Phonology Structure Grammar KPSG]

Korean Phonology Structure Grammar KPSG

[valid dictionary on the basis]

valid dictionary on the basis

[global behavior of a grammar]

global behavior of a grammar

[contrast to the usual approach]

contrast to the usual approach

[lexicon with 90 ,000 entries]

lexicon with 90 ,000 entries

[learning methodology applicable in general domains]

learning methodology applicable in general domains

[automatic generation of hypertext links]

automatic generation of hypertext links

[error characters]

error characters

[two distinct domains tutorial dialogue]

two distinct domains tutorial dialogue

[such questions]

such questions

[ONTOSCORE a system for scoring sets]

ONTOSCORE a system for scoring sets

[KANA-KANJI conversion]

KANA-KANJI conversion

[similar models]

similar models

[event elements]

event elements

[Minimalist grammars]

Minimalist grammars

[several aspects]

several aspects

[language parsing]

language parsing

[predicate-argument structures]

predicate-argument structures

[ATIS benchmark]

ATIS benchmark

[research context]

research context

[bigram features]

bigram features

[right context]

right context

[SL expressions]

SL expressions

[logical model]

logical model

[Natural Language Processing NLP systems]

Natural Language Processing NLP systems

[key features of the system]

key features of the system

[unigram model for machine translation]

unigram model for machine translation

[many systems]

many systems

[automatically identifying predicate argument structures]

automatically identifying predicate argument structures

[manual word]

manual word

[rule-based model]

rule-based model

[standard use]

standard use

[source text]

source texts, source text

[research projects]

research projects

[WSD accuracy of SMT models]

WSD accuracy of SMT models

[other texts]

other texts

[CCR formalism]

CCR formalism

[new theory of discourse structure]

new theory of discourse structure

[small corpus of the language]

small corpus of the language

[categorial grammar]

categorial grammar

[using a logical model in place]

using a logical model in place

[NLP techniques for document classification]

NLP techniques for document classification

[2-character Chinese names without title]

2-character Chinese names without title

[NIST score]

NIST score

[own knowledge]

own knowledge

[conventional word-trigram recognition requiring manual transcription]

conventional word-trigram recognition requiring manual transcription

[other approaches]

other approaches

[baseline methods]

baseline methods

[improving the performance of the word-sense]

improving the performance of the word-sense

[only resource]

only resources, only resource

[future needs]

future needs

[phrase translations from word-based alignments]

phrase translations from word-based alignments

[dialogue patterns]

dialogue patterns

[same sense]

same sense

[phonological knowledge]

phonological knowledge

[second model]

second model

[introduction to statistical machine translation]

introduction to statistical machine translation

[word fragments]

word fragments

[manual methods]

manual methods

[hard decisions]

hard decisions

[adequate tool]

adequate tool

[useful tool]

useful tool

[including a frequent word on APs]

including a frequent word on APs

[speaker leeway forming an utterance]

speaker leeway forming an utterance

[Montague semantics]

Montague semantics

[variation on the evaluation metric]

variation on the evaluation metric

[same problem]

same problem

[overlapping speech]

overlapping speech

[inherent ambiguity]

inherent ambiguity

[combination of the training speakers]

combination of the training speakers

[accuracy difference between the approaches]

accuracy difference between the approaches

[better predictions]

better predictions

[three tiers]

three tiers

[behavioral patterns]

behavioral patterns

[using an argumentation system by Konolige]

using an argumentation system by Konolige

[identifying hubs in an automaton]

identifying hubs in an automaton

[list of output sentence plans]

list of output sentence plans

[contextual information]

contextual information

[explicit use of preceding tag contexts]

explicit use of preceding tag contexts

[unification-based grammar]

unification-based grammar

[semantic-head-driven generation]

semantic-head-driven generation

[tagging LMs with confidence measures]

tagging LMs with confidence measures

[equivalence between the standard use]

equivalence between the standard use

[stop lists]

stop lists

[building polarity-tagged corpus from HTML documents]

building polarity-tagged corpus from HTML documents

[utterance features]

utterance features

[text structure]

text structure

[Graph unification]

Graph unification

[text corpus]

text corpus

[extraposition grammars]

extraposition grammars

[/Soames 1979/]

/Soames 1979/

[several approaches]

several approaches

[robust analysis]

robust analysis

[likely repair]

likely repair

[/Mercer 1987/]

/Mercer 1987/

[Begin/After tagging]

Begin/After tagging

[transfer component for parse reduction]

transfer component for parse reduction

[semantic set]

semantic set

[language L]

language L

[translates in another natural language]

translates in another natural language

[commonsense reasoning]

commonsense reasoning

[alternative index]

alternative index

[statistical machine translation tool kit]

statistical machine translation tool kit

[speech-search algorithm]

speech-search algorithm

[restricted-domain parsing]

restricted-domain parsing

[effect of the one sense]

effect of the one sense

[language understanding system with agents]

language understanding system with agents

[direct imitation of human performance]

direct imitation of human performance

[Bayesian learning]

Bayesian learning

[understanding process of the descriptions]

understanding process of the descriptions

[ambiguity handling]

ambiguity handling

[Transition Network]

Transition Network

[retrieval accuracy superior to any]

retrieval accuracy superior to any

[semantic questions]

semantic questions

[main dictionary]

main dictionary

[dependency relations]

dependency relations

[three different language system domains]

three different language system domains

[independent approach]

independent approach

[translation speed]

translation speed

[unsupervised word]

unsupervised word

[only system]

only system

[5 speed-up]

5 speed-up

[great effectiveness of this approach]

great effectiveness of this approach

[paraphrase sets]

paraphrase sets

[performance as a similarity measure]

performance as a similarity measure

[simple interpolation methods like log-linear interpolation]

simple interpolation methods like log-linear interpolation

[/Gazdar 1979/]

/Gazdar 1979/

[intelligent agent]

intelligent agents, intelligent agent

[computational system]

computational system, computational systems

[target texts]

target texts

[full analysis]

full analysis

[Thai prediction]

Thai prediction

[practical systems]

practical systems

[natural manner]

natural manner

[line length]

line length

[answering capability on free text]

answering capability on free text

[need developing polynomial time approximations]

need developing polynomial time approximations

[placing commas]

placing commas

[dictionary-based word]

dictionary-based word

[ideal aiming for complete interpretations]

ideal aiming for complete interpretations

[Arabic stemmer]

Arabic stemmer

[evaluation method]

evaluation method

[taking advantage of the symmetric patterns]

taking advantage of the symmetric patterns

[Multi-site ATIS Data COllection Working group]

Multi-site ATIS Data COllection Working group

[efficient decoder]

efficient decoder

[syntax-based constraint for word alignment]

syntax-based constraint for word alignment

[deterministic automaton]

deterministic automaton

[videodisc images]

videodisc images

[correct parse]

correct parse

[corresponding parser]

corresponding parser

[human-ranked sentence]

human-ranked sentence

[serious attention]

serious attention

[73 .2%]

73 .2%

[human cognition]

human cognition

[utility clustering]

utility clustering

[well-written discourse]

well-written discourse

[conversational agents]

conversational agents

[current systems]

current systems

[learning algorithm]

learning algorithm

[addressee of a dialogue act]

addressee of a dialogue act

[translation candidates]

translation candidates

[lack of manually sense-tagged data]

lack of manually sense-tagged data

[restrictions on the semantic categories]

restrictions on the semantic categories

[simple notations]

simple notations

[language model of abstract moves]

language model of abstract moves

[Results from experiments with substitution costs]

Results from experiments with substitution costs

[Such synthesis]

Such synthesis

[hot spots]

hot spots

[more traditional natural language interfaces]

more traditional natural language interfaces

[syntactic categories on the terminal]

syntactic categories on the terminal

[computational lexicon]

computational lexicon

[language generation]

language generation

[conversational features]

conversational features

[generative model]

generative model

[placing commas with a precision]

placing commas with a precision

[automatic methods]

automatic methods

[paraphrase sets from seed sentences]

paraphrase sets from seed sentences

[unified approach]

unified approach

[possible indexings]

possible indexings

[post-processing the output of OCR systems]

post-processing the output of OCR systems

[Markov probability of a correct chain]

Markov probability of a correct chain

[addition to an unsupervised component]

addition to an unsupervised component

[new problem]

new problem, new problems

[efficient Prolog cf query optimisation]

efficient Prolog cf query optimisation

[information sources]

information sources

[Statistical approach]

statistical approaches, Statistical approach

[hand-crafted rules]

hand-crafted rules

[adjoining words]

adjoining words

[real tasks]

real tasks

[resulting Arabic word segmentation system]

resulting Arabic word segmentation system

[nature of the missing fragments]

nature of the missing fragments

[finite-state models]

finite-state models

[manual determination]

manual determination

[social ranks of the referents]

social ranks of the referents

[erroneous chains]

erroneous chains

[knowledge for a user model]

knowledge for a user model

[Techniques for automatically training modules]

Techniques for automatically training modules

[world knowledge of the situation]

world knowledge of the situation

[usually the beginning of the sentence]

usually the beginning of the sentence

[new inputs]

new inputs

[potential antecedence of each element]

potential antecedence of each element

[using the NIST MT evaluation software]

using the NIST MT evaluation software

[conceptual problems]

conceptual problems

[combination with decision tree building]

combination with decision tree building

[conversational cues such as cue phrases]

conversational cues such as cue phrases

[/Soames 1982/]

/Soames 1982/

[linguistic knowledge]

linguistic knowledge

[word matchings]

word matchings

[35 .0%]

35 .0%

[human judges]

human judges

[method detecting]

method detecting

[including a variety of meta-inference types]

including a variety of meta-inference types

[equivalent words]

equivalent words

[linguistic generality]

linguistic generality

[eigenvectors of an adjacency graph's Laplacian]

eigenvectors of an adjacency graph's Laplacian

[principle advantage of this approach]

principle advantage of this approach

[,000 words]

,000 words

[particular representational format for meaning]

particular representational format for meaning

[annotating scheme]

annotating scheme

[preliminary analysis]

preliminary analysis

[logical definition of Minimalist grammars]

logical definition of Minimalist grammars

[English sentence]

English sentences, English sentence

[two ways]

two ways

[research using machine learning techniques]

research using machine learning techniques

[candidate antecedents]

candidate antecedents

[logical definition]

logical definition

[lateral directions]

lateral directions

[syntactic structure]

syntactic structure

[correlation between automatic evaluation measures]

correlation between automatic evaluation measures

[Paul a computer text generation system]

Paul a computer text generation system

[linguistic objects]

linguistic objects

[event networks]

event networks

[fully-connected first-order statistical class grammar]

fully-connected first-order statistical class grammar

[superordinate substitution]

superordinate substitution

[language learning]

language learning

[NE performance for some NE types]

NE performance for some NE types

[several kinds of linguistic knowledge]

several kinds of linguistic knowledge

[statistical systems]

statistical systems

[linguistic pattern]

linguistic pattern

[bi-directional fragment]

bi-directional fragment

[describing inflection for the purpose]

describing inflection for the purpose

[special method]

special method

[approach inducing a grammar from corpora]

approach inducing a grammar from corpora

[state-of-the-art approaches]

state-of-the-art approaches

[surprising finding]

surprising finding

[Canadian Hansards]

Canadian Hansards

[evaluating word sense disambiguation WSD models]

evaluating word sense disambiguation WSD models

[coherent (given]

coherent (given

[development of a message-passing infrastructure]

development of a message-passing infrastructure

[Conceptual Dependency]

Conceptual Dependency

[recent advances in phrasal translation]

recent advances in phrasal translation

[subject of much recent debate]

subject of much recent debate

[avoidance of copying of unmodified subgraphs]

avoidance of copying of unmodified subgraphs

[including vocabulary]

including vocabulary

[probably-correct rules]

probably-correct rules

[perplexing fact]

perplexing fact

[centering on the construction of models]

centering on the construction of models

[full meaning]

full meaning

[preliminary experiments]

preliminary experiments

[fuzzy syllogism]

fuzzy syllogism

[similar indifference]

similar indifference

[information redundancy in multilingual input]

information redundancy in multilingual input

[system combination for unsupervised WSD]

system combination for unsupervised WSD

[information nugget]

information nugget

[expressions including a frequent word]

expressions including a frequent word

[Resource Management]

Resource Management

[syntax-based decoder]

syntax-based decoder

[important information]

important information

[annotation experiment]

annotation experiment

[most important task building the editor]

most important task building the editor

[correcting these errors in Japanese bunsetsu]

correcting these errors in Japanese bunsetsu

[semantic verb]

semantic verb

[standard grammar]

standard grammar

[parse]

parses, parse

[training corpus]

training corpus

[understanding system]

understanding system

[example-retrieval ER]

example-retrieval ER

[primarily passing some computational issues]

primarily passing some computational issues

[cohesion constraint]

cohesion constraint

[traditional approach]

traditional approach, traditional approaches

[multiple output]

multiple output

[asynchronous message passing between actors]

asynchronous message passing between actors

[combining knowledge about dialog schemata]

combining knowledge about dialog schemata

[system RAREAS]

system RAREAS

[strongly biasing]

strongly biasing

[three separate components the structure]

three separate components the structure

[separation margin between the candidate]

separation margin between the candidate

[proper substitutions for these elements]

proper substitutions for these elements

[best confidence]

best confidence

[thus the number of edges]

thus the number of edges

[hand-crafted system]

hand-crafted system

[superior performance]

superior performance

[synchronous dependency]

synchronous dependency

[structures containing sets of features]

structures containing sets of features

[Robust parsing]

Robust parsing

[clustering algorithm]

clustering algorithm

[use of multiple parsing strategies]

use of multiple parsing strategies

[corrective training]

corrective training

[word-based models]

word-based models

[maintaining the quasi-destructive scheme's ability]

maintaining the quasi-destructive scheme's ability

[original document]

original document

[anaphora references]

anaphora references

[semantic annotation]

semantic annotation

[experiment using machine translation output]

experiment using machine translation output

[terms of event type networks]

terms of event type networks

[identifying systematic patterns in translation data]

identifying systematic patterns in translation data

[number of performance enhancing techniques]

number of performance enhancing techniques

[computational phonological system speech recognition]

computational phonological system speech recognition

[information related to a particular scenario]

information related to a particular scenario

[advantage of detailed linguistic information]

advantage of detailed linguistic information

[conversion rules]

conversion rules

[hard decisions using the reference]

hard decisions using the reference

[experimental evaluation of summarization quality]

experimental evaluation of summarization quality

[kanji-kana characters]

kanji-kana characters

[consisting of 126 ,610 sentences]

consisting of 126 ,610 sentences

[application at the character level]

application at the character level

[conversational speech]

conversational speech

[IE paradigm]

IE paradigm

[two strategies for language understanding]

two strategies for language understanding

[Isomorphic Grammars approach to MT]

Isomorphic Grammars approach to MT

[Non Verbal]

Non Verbal

[inducing a grammar from parallel corpora]

inducing a grammar from parallel corpora

[query optimisation in a relational database]

query optimisation in a relational database

[unsupervised learning algorithm for identification]

unsupervised learning algorithm for identification

[speculative claim]

speculative claim, speculative claims

[Dynamic Hierarchical Phrasal Lexicon DHPL]

Dynamic Hierarchical Phrasal Lexicon DHPL

[English-Chinese bitexts]

English-Chinese bitexts

[retrieval performance]

retrieval performance

[good insight into the potential]

good insight into the potential

[cooperative responses to each user]

cooperative responses to each user

[computational analysis of move structures]

computational analysis of move structures

[word string with the performance]

word string with the performance

[valuable task]

valuable task

[one sentences]

one sentences

[parser incorporating the control structure]

parser incorporating the control structure

[linguistic analysis]

linguistic analysis

[new stems]

new stems

[syntactic information]

syntactic information

[algorithmic additions]

algorithmic additions

[combining the results from answering agents]

combining the results from answering agents

[automatically evaluating the summarization quality]

automatically evaluating the summarization quality

[chart-based phrase]

chart-based phrase

[probabilistic model]

probabilistic model

[transfer phase in machine translation]

transfer phase in machine translation

[ambiguous sentences]

ambiguous sentences

[new strategy]

new strategy

[applying statistical models to data]

applying statistical models to data

[execution at the appropriate database]

execution at the appropriate database

[speech understanding]

speech understanding

[ranking rules]

ranking rules

[source language]

source language

[large Chinese non-parallel newspaper corpora]

large Chinese non-parallel newspaper corpora

[effectiveness of answer resolution algorithm]

effectiveness of answer resolution algorithm

[dialog schemata]

dialog schemata

[further work]

further work

[training phase]

training phase

[SUN Sparcstations]

SUN Sparcstations

[text processing]

text processing

[different amounts]

different amounts

[best existing]

best existing

[successive learners]

successive learners

[quadratic time]

quadratic time

[specialized editor]

specialized editor

[significant improvement in alignment quality]

significant improvement in alignment quality

[sentence structure]

sentence structure

[practical considerations]

practical considerations

[semantically classifying undisambiguated SCF data]

semantically classifying undisambiguated SCF data

[49 .3%]

49 .3%

[processing description]

processing description

[using parsing strategies in a fashion]

using parsing strategies in a fashion

[general mechanism]

general mechanism

[world knowledge]

world knowledge

[CCLINC Common Coalition Language System]

CCLINC Common Coalition Language System

[novel method building polarity-tagged corpus]

novel method building polarity-tagged corpus

[decoding algorithm]

decoding algorithm

[boosting approach]

boosting approach

[dramatic improvements]

dramatic improvements

[using a binary relational calculus notation]

using a binary relational calculus notation

[Korean-to-English machine]

Korean-to-English machine

[much more commonly available resource]

much more commonly available resource

[unsupervised algorithm]

unsupervised algorithm

[place of a denotational semantics]

place of a denotational semantics

[text collection]

text collections, text collection

[feature space]

feature space

[Basic methodology]

Basic methodology

[segment order-sensitive methods in terms]

segment order-sensitive methods in terms

[2 SRHs]

2 SRHs

[common ground in human-computer interaction]

common ground in human-computer interaction

[word-aligned corpora]

word-aligned corpora

[disambiguation the interpretation of ellipses]

disambiguation the interpretation of ellipses

[hand-produced sets]

hand-produced sets

[sets of individually high-precision rules]

sets of individually high-precision rules

[two new word-sense disambiguation systems]

two new word-sense disambiguation systems

[FERRET an interactive question-answering Q/A system]

FERRET an interactive question-answering Q/A system

[user satisfaction]

user satisfaction

[Speech Recognition]

Speech Recognition

[accurate translation]

accurate translation

[kind of Semantic Role System]

kind of Semantic Role System

[ambiguity packing]

ambiguity packing

[English-Chinese bitexts of the laws]

English-Chinese bitexts of the laws

[High quality]

High quality, high quality

[pronominal anaphora]

pronominal anaphora

[core technology]

core technology

[adaptive nature]

adaptive nature

[natural grammar]

natural grammar

[Chart-like parsing]

Chart-like parsing

[state of focus of attention]

state of focus of attention

[OCR system]

OCR system, OCR systems

[decision tree]

decision tree

[logical formalism]

logical formalism

[context-free grammars]

context-free grammars

[reformulation of the erroneous sentence]

reformulation of the erroneous sentence

[requiring significantly less development time]

requiring significantly less development time

[VME bus]

VME bus

[synthesis system]

synthesis system

[development of machine translation systems]

development of machine translation systems

[sentence analysis]

sentence analysis, sentence analyses

[Much effort]

Much effort

[calculating eigenvectors]

calculating eigenvectors

[best system]

best system

[Computer programs]

Computer programs

[employing a traditional generative phonological approach]

employing a traditional generative phonological approach

[available knowledge]

available knowledge

[hard problems]

hard problems

[state-of-the-art performance]

state-of-the-art performance

[Finite State]

Finite State

[comparative evaluations]

comparative evaluations

[unidentified portion]

unidentified portion

[training material]

training material

[using these tree-based models in combination]

using these tree-based models in combination

[sense coverage]

sense coverage

[decision list]

decision list

[usually in a rightward manner]

usually in a rightward manner

[last decade]

last decades, last decade

[HBG model]

HBG model

[most birds]

most birds

[gathering statistics on the output]

gathering statistics on the output

[official rankings]

official rankings

[demonstrative forms]

demonstrative forms

[unsupervised training]

unsupervised training

[respect to the original question]

respect to the original question

[character recognition]

character recognition

[high cost]

high costs, high cost

[automating certain aspects of language learning]

automating certain aspects of language learning

[on-line communities]

on-line communities

[input text]

input text

[automatic word]

automatic word

[non-literal aspects]

non-literal aspects

[terms of the inference types]

terms of the inference types

[Finding the language for a need]

Finding the language for a need

[optimum configuration]

optimum configuration

[event terms]

event terms

[local trees]

local trees

[preference knowledge]

preference knowledge

[conjunction with a LR parser]

conjunction with a LR parser

[LHIP a system for grammar development]

LHIP a system for grammar development

[improvement on several machine translation tests]

improvement on several machine translation tests

[new mechanism]

new mechanism

[model using associative processors APs]

model using associative processors APs

[zero occurrences of a morpheme]

zero occurrences of a morpheme

[addressee identification in four-participants meetings]

addressee identification in four-participants meetings

[clear evidence from the input]

clear evidence from the input

[trigram model]

trigram models, trigram model

[assignment of descriptors from phrases]

assignment of descriptors from phrases

[essential underpinning]

essential underpinning

[different strategies]

different strategies

[Deictic reference about the discourse]

Deictic reference about the discourse

[requirements for a software infrastructure]

requirements for a software infrastructure

[paraphrase extraction]

paraphrase extraction

[essential roles]

essential roles

[semantic grammar]

semantic grammar

[Interpreting metaphors]

Interpreting metaphors

[lexical information]

lexical information

[robust recognition of extra-grammatical input]

robust recognition of extra-grammatical input

[novel likewise automatic evaluation method]

novel likewise automatic evaluation method

[full meaning of the sentence]

full meaning of the sentence

[implicit quantifiers]

implicit quantifiers

[syntactic ambiguities]

syntactic ambiguities

[STTK a machine translation tool kit]

STTK a machine translation tool kit

[standard ATN]

standard ATN

[problem of the automatic detection]

problem of the automatic detection

[building a language model of moves]

building a language model of moves

[test corpus]

test corpus

[target sentence]

target sentence

[new type of transfer system]

new type of transfer system

[restrictive statements]

restrictive statements

[annealing approach]

annealing approach

[key prediction]

key prediction

[including part of speech tagging]

including part of speech tagging

[error correction of ill-formed input]

error correction of ill-formed input

[international collaborations]

international collaborations

[sentence scoping]

sentence scoping

[operational semantics]

operational semantics

[computational analysis]

computational analysis

[two approaches]

two approaches

[logical form]

logical form

[use by Language Processing systems]

use by Language Processing systems

[multilingual data]

multilingual data

[results of a practical evaluation]

results of a practical evaluation

[Using LCS-Marine]

Using LCS-Marine

[naturally sounding dialog systems within reach]

naturally sounding dialog systems within reach

[current capabilities in dialogue systems]

current capabilities in dialogue systems

[entity-oriented parser]

entity-oriented parser

[additional features]

additional features

[abductive inference]

abductive inference

[88 .2%]

88 .2%

[fragmentary recognition]

fragmentary recognition

[priors in conditional loglinear models]

priors in conditional loglinear models

[meaning of a proposition p]

meaning of a proposition p

[7% improvement in recognition accuracy]

7% improvement in recognition accuracy

[applying standard MT evaluation methods (BLEU]

applying standard MT evaluation methods (BLEU

[new method]

new methods, new method

[discourse understanding]

discourse understanding

[new technique]

new technique

[training algorithm]

training algorithm

[computation model]

computation model

[pragmatics processing]

pragmatics processing

[efficient algorithm]

efficient algorithm

[light of various rhetorical functions]

light of various rhetorical functions

[problem automatically predicting segment boundaries]

problem automatically predicting segment boundaries

[reestimation formulas]

reestimation formulas

[40 utterances from the target speaker]

40 utterances from the target speaker

[practical ways dealing with complexity]

practical ways dealing with complexity

[shallow linguistic features of questions]

shallow linguistic features of questions

[real world]

real world

[resulting tagger]

resulting tagger

[parsing a set of phrases]

parsing a set of phrases

[cooccurrence-based word]

cooccurrence-based word

[calculus notation]

calculus notation

[global scene]

global scene

[also graphs indispensable to papers]

also graphs indispensable to papers

[limitations in comparison with WSD models]

limitations in comparison with WSD models

[comma checker]

comma checker

[parsing data]

parsing data

[information request]

information request

[patterns in machine translation output]

patterns in machine translation output

[first experiment in a series]

first experiment in a series

[second requirement]

second requirement

[practical techniques]

practical techniques

[evaluating a representative Chinese-to-English SMT model]

evaluating a representative Chinese-to-English SMT model

[nonverbal behaviors]

nonverbal behaviors

[semantic coherence]

semantic coherence

[good guidance for novice users]

good guidance for novice users

[clustering subcategorization frame SCF distributions]

clustering subcategorization frame SCF distributions

[block reordering]

block reordering

[best condition]

best condition

[efficacy applying automated evaluation techniques]

efficacy applying automated evaluation techniques

[building a non-English Arabic stemmer]

building a non-English Arabic stemmer

[defeat rules]

defeat rules

[formal analysis]

formal analysis

[tactical personnel]

tactical personnel

[Byron's system]

Byron's system

[accurate results]

accurate results

[first utterance]

first utterance

[several dimensions]

several dimensions

[parallel text]

parallel text

[head nods]

head nods

[similar approach]

similar approach

[competing members]

competing members

[overall pattern]

overall pattern

[global meaning of a sentence]

global meaning of a sentence

[text-to-speech form]

text-to-speech form

[redundant copying]

redundant copying

[human users]

human users

[straight C]

straight C

[decision how]

decision how

[compound noun]

compound nouns, compound noun

[manual transcription]

manual transcription

[Shieber [11]]

Shieber [11]

[commercial systems outputting unsegmented texts]

commercial systems outputting unsegmented texts

[approximation error]

approximation error

[accuracy rate of syntactic disambiguation]

accuracy rate of syntactic disambiguation

[Word Identification]

Word Identification

[underlying theory]

underlying theory

[Error analysis]

Error analysis

[one thing]

one thing

[Arabic corpus]

Arabic corpus

[basic idea of this method]

basic idea of this method

[specific information]

specific information

[category transitions for speech repairs]

category transitions for speech repairs

[FlexP a bottom-up pattern-matching parser]

FlexP a bottom-up pattern-matching parser

[basic mapping]

basic mapping

[Kyoto city bus information system]

Kyoto city bus information system

[even in a precise way]

even in a precise way

[Listen-Communicate-Show LCS]

Listen-Communicate-Show LCS

[human interaction]

human interaction

[logical well-formedness conditions on trees]

logical well-formedness conditions on trees

[automatically evaluating answers to definition questions]

automatically evaluating answers to definition questions

[part of industry watch function]

part of industry watch function

[new view for extensional reference]

new view for extensional reference

[representing determiners]

representing determiners

[fundamental concern]

fundamental concern

[lexical information from the parse tree]

lexical information from the parse tree

[initial ranking]

initial ranking

[Complete lexical distribution of grammatical knowledge]

Complete lexical distribution of grammatical knowledge

[robust mixed-initiative speech dialogue interactions]

robust mixed-initiative speech dialogue interactions

[kernel function]

kernel function, kernel functions

[new algorithm]

new algorithm

[need for answers to questions]

need for answers to questions

[conventional treatments]

conventional treatments

[working in a 'synchronous' way]

working in a 'synchronous' way

[Chinese word]

Chinese word

[human annotators]

human annotators

[whole sentence]

whole sentence

[discourse segments]

discourse segments

[identical information]

identical information

[cue phrases]

cue phrases

[disambiguation algorithms]

disambiguation algorithms

[increasing worst-case parsing time complexity]

increasing worst-case parsing time complexity

[discourse processing]

discourse processing, Discourse processing

[disambiguation process]

disambiguation process

[English stemmer]

English stemmer

[naturally aggregate]

naturally aggregate

[phrase-based statistical machine translation method]

phrase-based statistical machine translation method

[entity-oriented approach to restricted-domain parsing]

entity-oriented approach to restricted-domain parsing

[four applications]

four applications

[expert translation]

expert translations, expert translation

[Link detection]

Link detection

[conventional methods]

conventional methods

[manual transcription of training data]

manual transcription of training data

[multiple candidates for the understanding result]

multiple candidates for the understanding result

[presuppositional nature]

presuppositional nature

[knowledge concerning translation equivalence of expressions]

knowledge concerning translation equivalence of expressions

[Similar results]

Similar results

[around 97% exact match accuracy]

around 97% exact match accuracy

[direction-giving task]

direction-giving task

[Communicator participants]

Communicator participants

[diverse problems]

diverse problems

[little robustness]

little robustness

[instead of the traditional practice]

instead of the traditional practice

[large class]

large class

[Dynamic Grammars]

Dynamic Grammars

[bigger corpus]

bigger corpus

[computer program]

computer program

[separate modules]

separate modules

[unsupervised component]

unsupervised component

[little speech]

little speech

[similar fashion]

similar fashion

[utility applying MT evaluation methods]

utility applying MT evaluation methods

[newspaper stories]

newspaper stories

[lexical-cohesion features]

lexical-cohesion features

[syntax-based statistical machine translation system]

syntax-based statistical machine translation system

[CFG filtering techniques for LTAG]

CFG filtering techniques for LTAG

[output quality]

output quality

[end-to-end process]

end-to-end process

[resource-frugal approach]

resource-frugal approach

[everyday rejoinders]

everyday rejoinders

[discourse structure]

discourse structure

[important role conveying the meaning]

important role conveying the meaning

[method attaining a design overheads]

method attaining a design overheads

[program RINA]

program RINA

[telephone dialogue]

telephone dialogue

[scruffy texts in the domain]

scruffy texts in the domain

[exploiting a large non-parallel corpus]

exploiting a large non-parallel corpus

[monolingual corpus]

monolingual corpus, monolingual corpora

[PageRank algorithm]

PageRank algorithm

[possibility using the language model]

possibility using the language model

[referents anaphora]

referents anaphora

[potential antecedence]

potential antecedence

[syntactic expectations]

syntactic expectations

[concrete examples]

concrete examples

[two experiments]

two experiments

[tracking the discourse through the operation]

tracking the discourse through the operation

[boosting method]

boosting method

[robust parser]

robust parser

[including the influence of training factors]

including the influence of training factors

[semantic categories]

semantic categories

[State-of-the-art Question Answering QA systems]

State-of-the-art Question Answering QA systems

[natural language texts e .g]

natural language texts e .g

[maximum entropy word alignment algorithm]

maximum entropy word alignment algorithm

[February 1992]

February 1992

[automatically training SPoT on the basis]

automatically training SPoT on the basis

[13% decrease in F-measure error]

13% decrease in F-measure error

[previous studies]

previous studies

[to reasoning]

to reasoning

[SI result]

SI result

[conversational context]

conversational context

[speaker adaptation]

speaker adaptation

[problems of vagueness in annotation]

problems of vagueness in annotation

[Two styles]

Two styles, two styles

[erroneous sentence]

erroneous sentence

[document processing]

document processing

[instantiation of the one sense]

instantiation of the one sense

[automatic evaluation]

automatic evaluation

[correctness of answers to questions]

correctness of answers to questions

[Prolog form]

Prolog form

[Arabic word]

Arabic word

[statistical class]

statistical class

[IBM Models]

IBM Models, IBM models

[trainable components]

trainable components

[97 .24%]

97 .24%

[understanding system with intelligent mobile agents]

understanding system with intelligent mobile agents

[English questions]

English questions

[adopting a simple MT-based paraphrasing technique]

adopting a simple MT-based paraphrasing technique

[using a bigger a corpus]

using a bigger a corpus

[Boolean conditions on the cooccurrence]

Boolean conditions on the cooccurrence

[Experiment results on ACE corpora]

Experiment results on ACE corpora

[Sources of training data suitable]

Sources of training data suitable

[heuristic rules]

heuristic rules

[browsing the repository of data objects]

browsing the repository of data objects

[real-world environments]

real-world environments

[errors in the machine translations]

errors in the machine translations

[maximum-entropy model]

maximum-entropy models, maximum-entropy model

[2 inductive decision tree learning]

2 inductive decision tree learning

[kind of similarity between words]

kind of similarity between words

[parsing-as-deduction in a resource logic]

parsing-as-deduction in a resource logic

[Japanese-English translation]

Japanese-English translation

[extensional reference]

extensional reference

[semantically coherent speech recognition hypotheses]

semantically coherent speech recognition hypotheses

[paraphrase probability]

paraphrase probability

[trigram language]

trigram language

[minimal DFA]

minimal DFA

[graphical model]

graphical model

[fuzzy quantifier]

fuzzy quantifiers, fuzzy quantifier

[dominance graphs]

dominance graphs

[language modeling of conversational speech]

language modeling of conversational speech

[grammatical formalisms]

grammatical formalisms

[current sentence]

current sentence

[12,000 utterances]

12,000 utterances

[relationships among the target variables]

relationships among the target variables

[relevant approach]

relevant approach

[searching for answers in multiple corpora]

searching for answers in multiple corpora

[language presuppositions]

language presuppositions

[first learner]

first learner

[edit operation]

edit operation

[stemming model]

stemming model

[logical formulas]

logical formulas

[lesser extent]

lesser extent

[certainty conditions]

certainty conditions

[theoretical work]

theoretical work

[new measure]

new measure

[exhaustive comparison]

exhaustive comparison

[great benefit to language pairs]

great benefit to language pairs

[theoretical literature]

theoretical literature

[English coordinate structure analysis model]

English coordinate structure analysis model

[short time]

short time

[recent work]

recent work

[useful features for paraphrase classification]

useful features for paraphrase classification

[vagueness of the spatial concepts]

vagueness of the spatial concepts

[polynomial time solution for any]

polynomial time solution for any

[many speakers]

many speakers

[interactively visualizing]

interactively visualizing

[early copying]

early copying

[case-based reasoning in machine translation]

case-based reasoning in machine translation

[sensitive logic]

sensitive logic

[blackboard-like architecture]

blackboard-like architecture

[neat texts]

neat texts

[paraphrase classification]

paraphrase classification

[thereby reducing the amount of ambiguity]

thereby reducing the amount of ambiguity

[word-level alignment]

word-level alignment

[spectral approach]

spectral approach

[out-of-domain corpus]

out-of-domain corpus

[parsing technique]

parsing technique

[F-measure error]

F-measure error

[/Soames 1982/]

/Soames 1982/

[training examples]

training examples

[overview of the CSR corpus]

overview of the CSR corpus

[analysis step]

analysis step

[Czech-English statistical machine translation system]

Czech-English statistical machine translation system

[broad domains]

broad domains

[image understanding with text understanding]

image understanding with text understanding

[advantages of the three strategies]

advantages of the three strategies

[textual representation]

textual representation

[translingual reach into other languages]

translingual reach into other languages

[verb group]

verb group

[fake symbol]

fake symbol

[unification algorithms avoidance of copying]

unification algorithms avoidance of copying

[several voting-]

several voting-

[translation into an equivalent RCG]

translation into an equivalent RCG

[storage media]

storage media

[binary features]

binary features

[large-scale grammars]

large-scale grammars

[using m-th order Markov chain model]

using m-th order Markov chain model

[combination with a clustering process]

combination with a clustering process

[substitution costs]

substitution costs

[same beliefs]

same beliefs

[monitoring of lack of feedback]

monitoring of lack of feedback

[Schutze's idea of evaluation algorithms]

Schutze's idea of evaluation algorithms

[single system]

single system

[sentence-level correlation]

sentence-level correlation

[unsegmented corpus]

unsegmented corpus

[adding linguistic annotation to corpora]

adding linguistic annotation to corpora

[Grolier's Encyclopedia]

Grolier's Encyclopedia

[other languages]

other languages

[part-of-speech-based criterion]

part-of-speech-based criterion

[target document]

target document

[domain-independent lexica]

domain-independent lexica

[benchmark system]

benchmark system

[image understanding]

image understanding

[error reduction of 4 .4%]

error reduction of 4 .4%

[valuable lexicons]

valuable lexicons

[human interaction with data sources]

human interaction with data sources

[using commands]

using commands

[other tools]

other tools

[top-down derivation]

top-down derivation

[evaluation techniques]

evaluation techniques

[challenges integrating automatic Q/A applications]

challenges integrating automatic Q/A applications

[describing a case study avoiding failures]

describing a case study avoiding failures

[reference sets]

reference sets

[conditional models]

conditional models

[Sentence planning]

Sentence planning

[Language understanding work at Paramax]

Language understanding work at Paramax

[more complex mixtures of techniques]

more complex mixtures of techniques

[multiple translations]

multiple translations

[use of various lexical features]

use of various lexical features

[translation relation]

translation relation

[translation tool]

translation tool

[negative feedback]

negative feedback

[tree-to-tree translation of dependency structures]

tree-to-tree translation of dependency structures

[particular domain]

particular domain

[dialogue acquisition]

dialogue acquisition

[investigating the combinatorics of free indexation]

investigating the combinatorics of free indexation

[final language with overt case markers]

final language with overt case markers

[several respects]

several respects

[elastic constraints]

elastic constraints

[Several extensions of this basic idea]

Several extensions of this basic idea

[missing periods]

missing periods

[polysemous word]

polysemous words, polysemous word

[role of purpose in discourse]

role of purpose in discourse

[specific domain of HK laws]

specific domain of HK laws

[semantic network]

semantic networks, semantic network

[practical efficiency of RCL parsers]

practical efficiency of RCL parsers

[correct parse of a sentence]

correct parse of a sentence

[classification accuracy]

classification accuracy

[indexing according to simple character bigrams]

indexing according to simple character bigrams

[many languages]

many languages

[pilot system]

pilot system

[b perspective-taking]

b perspective-taking

[useful insights]

useful insights

[new edge]

new edge

[parsing the Wall Street Journal treebank]

parsing the Wall Street Journal treebank

[integrating language understanding with speech recognition]

integrating language understanding with speech recognition

[strongly biasing parsing toward meanings]

strongly biasing parsing toward meanings

[study on communications between APs]

study on communications between APs

[string with the best performance (typically]

string with the best performance (typically

[picking the best hypothesis corresponding]

picking the best hypothesis corresponding

[Isomorphic Grammars for Machine Translation]

Isomorphic Grammars for Machine Translation

[means of a beam-search decoder]

means of a beam-search decoder

[using a set of word alignments]

using a set of word alignments

[French sentence]

French sentence

[prototype system]

prototype system

[young men]

young men

[bilingual material the Canadian Hansards]

bilingual material the Canadian Hansards

[affix lists]

affix lists

[conveying the meaning of an utterance]

conveying the meaning of an utterance

[exploratory corpus analysis accessible to researchers]

exploratory corpus analysis accessible to researchers

[little inflection]

little inflection

[intensional concepts]

intensional concepts

[event]

events, event

[English grammar]

English grammar

[6 .5%]

6 .5%

[reference model]

reference model

[more robust---an agenda-based scheduling parser]

more robust---an agenda-based scheduling parser

[showing the WSD accuracy significantly lower]

showing the WSD accuracy significantly lower

[computer system]

computer system

[little inflection such as English]

little inflection such as English

[model TDMT Transfer-Driven Machine Translation]

model TDMT Transfer-Driven Machine Translation

[speech information of the words]

speech information of the words

[Such mistakes]

Such mistakes

[experience planning COMPLEX a computational lexicon]

experience planning COMPLEX a computational lexicon

[such systems]

such systems

[identifying a predicate with honorifics]

identifying a predicate with honorifics

[output summary]

output summary

[learning approach]

learning approach

[content-independent formalism]

content-independent formalism

[LR parsing]

LR parsing

[pre-tagged corpus]

pre-tagged corpus

[two times]

two times

[cyclic structures]

cyclic structures

[expressions of natural languages to semantics]

expressions of natural languages to semantics

[pronoun resolution]

pronoun resolution

[ensemble methods in machine learning]

ensemble methods in machine learning

[test-score semantics]

test-score semantics

[higher models]

higher models

[typical usage]

typical usage

[developers of machine translation systems]

developers of machine translation systems

[Gaussian mixture]

Gaussian mixture

[same time]

same time

[WSD accuracy of SMT models]

WSD accuracy of SMT models

[little difficulty]

little difficulty

[objective machine translation evaluation measures]

objective machine translation evaluation measures

[linguistic domain]

linguistic domain

[increasing the dialogue duration for users]

increasing the dialogue duration for users

[good performance]

good performance, best performance

[Semantic information]

Semantic information

[synchronous grammars]

synchronous grammars

[source code]

source code

[implications for current discourse processing algorithms]

implications for current discourse processing algorithms

[three types of the errors]

three types of the errors

[lexical hierarchy]

lexical hierarchy

[obviating the need for answers]

obviating the need for answers

[yielding a treatment of Montague semantics]

yielding a treatment of Montague semantics

[BLEU at the character level]

BLEU at the character level

[ranking methods]

ranking methods

[BLEU score]

BLEU scores, BLEU score

[algorithmic efficiency]

algorithmic efficiency

[building spelling-checkers for such languages]

building spelling-checkers for such languages

[speed of the resulting program]

speed of the resulting program

[RCL parser]

RCL parsers, RCL parser

[large database]

large database, larger database

[program APT]

program APT

[cooperative responses]

cooperative responses

[unigram counts]

unigram counts

[lexical paraphrases]

lexical paraphrases

[language acquisition]

language acquisition

[significant improvements over the WI systems]

significant improvements over the WI systems

[non-overlapping intervals]

non-overlapping intervals

[syntactic paraphrases]

syntactic paraphrases

[extra-grammatical input]

extra-grammatical input

[parser evaluation]

parser evaluation

[elementary structures]

elementary structures

[phrase length]

phrase length

[temporary belief about the world]

temporary belief about the world

[saturation of the final structure]

saturation of the final structure

[apparent tendency]

apparent tendency

[Stabler's formalization of Chomsky's minimalist program]

Stabler's formalization of Chomsky's minimalist program

[1 an set of features]

1 an set of features

[multi-site evaluation]

multi-site evaluation

[similar performance]

similar performance

[ASR output]

ASR output

[computational model of the process]

computational model of the process

[content understanding of the document]

content understanding of the document

[set of manual word alignments]

set of manual word alignments

[training method]

training method

[new methodology automatically training SPoT]

new methodology automatically training SPoT

[metaphor interpretation]

metaphor interpretation

[stereotypic dialogue]

stereotypic dialogue

[October '91]

October '91

[information need]

information need

[directly on word sense disambiguation performance]

directly on word sense disambiguation performance

[fundamental problem in user modeling]

fundamental problem in user modeling

[novel way]

novel way

[Japanese bunsetsu]

Japanese bunsetsu

[outstanding problems with the approach]

outstanding problems with the approach

[10K sentences]

10K sentences

[DARPA speech recognition technology development]

DARPA speech recognition technology development

[attentional focus]

attentional focus

[evaluation metric]

evaluation metric, evaluation metrics

[important role in the determination]

important role in the determination

[multiple ways]

multiple ways

[practical evaluation]

practical evaluation

[research program]

research program

[obvious implementation]

obvious implementation

[culling information from a large inflow]

culling information from a large inflow

[top-ranked plan]

top-ranked plan

[EBMT system]

EBMT system

[Overall summarization quality of the system]

Overall summarization quality of the system

[speaker's utterance]

speaker's utterance

[SUMMIT recognizer]

SUMMIT recognizer

[young men like young women]

young men like young women

[correct candidates for the errors]

correct candidates for the errors

[Dialogue strategies]

Dialogue strategies

[linguistic representation]

linguistic representation

[search space]

search space

[parsing complexity]

parsing complexity

[people's answers to everyday questions]

people's answers to everyday questions

[system FROFF]

system FROFF

[multi engine machine translation system]

multi engine machine translation system

[everyday questions]

everyday questions

[log-linear block]

log-linear block

[target variables]

target variables

[feature selection methods within models]

feature selection methods within models

[lexical generalizations]

lexical generalizations

[systematic patterns]

systematic patterns

[linguistic segments]

linguistic segments

[rule-reduction algorithm applying mutual information]

rule-reduction algorithm applying mutual information

[theoretical accounts]

theoretical accounts

[Korean newspaper articles on missiles]

Korean newspaper articles on missiles

[DoPS system]

DoPS system

[IDUS Intelligent Document Understanding System]

IDUS Intelligent Document Understanding System

[segments of actual tape-recorded descriptions]

segments of actual tape-recorded descriptions

[sentence-plan-ranker SPR]

sentence-plan-ranker SPR

[component performance]

component performance

[using standard WSD evaluation methodology]

using standard WSD evaluation methodology

[routine basis]

routine basis

[many theories of anaphora resolution]

many theories of anaphora resolution

[classification accuracy of the method]

classification accuracy of the method

[still satisfying human communication needs]

still satisfying human communication needs

[Annotating honorifics]

Annotating honorifics

[formalism's usage]

formalism's usage

[cluster number estimation on the eigenvectors]

cluster number estimation on the eigenvectors

[basic idea]

basic idea

[subparagraph level]

subparagraph level

[identifying unknown words in Chinese newspapers]

identifying unknown words in Chinese newspapers

[usual approach]

usual approach

[language identification]

language identification

[providing an interpretation for values]

providing an interpretation for values

[text browser]

text browser, text browsers

[cooccurrence statistics]

cooccurrence statistics

[elaborate documents]

elaborate documents

[reproducability of a gold standard]

reproducability of a gold standard

[problem enumerating all possible indexings]

problem enumerating all possible indexings

[declarative way]

declarative way

[unsupervised WSD]

unsupervised WSD

[using a phrase in another language]

using a phrase in another language

[constraint-based parser/generator]

constraint-based parser/generator

[monolingual UCG]

monolingual UCG

[system's output]

system's output

[Japanese texts]

Japanese texts

[syntax proper]

syntax proper

[link decisions]

link decisions

[own width]

own width

[original model]

original model

[Previous research]

Previous research

[segmentation accuracy]

segmentation accuracy

[workability of the general mechanism]

workability of the general mechanism

[people's answers]

people's answers

[individual phrases]

individual phrases

[dependency structure]

dependency structure, dependency structures

[inheritance mechanisms]

inheritance mechanisms

[constraint]

constraints, constraint

[sense]

sense, senses

[segment]

segments, segment

[tree]

trees, tree, Tree

[role]

role, roles, Role

[number]

number

[alignment]

alignment, alignments

[dialog]

dialog, dialogs

[source]

Source, source, Sources

[character]

character, characters

[including]

including

[expression]

expressions, Expressions, expression

[combination]

combination, Combination

[advantage]

advantage, advantages

[candidate]

candidate, candidates

[semantics]

semantics

[ranking]

ranks, ranking

[purpose]

purposes, purpose

[recognition]

recognition

[tool]

tool, tools

[accuracy]

accuracy

[pattern]

patterns, pattern

[formalism]

formalism, formalisms

[database]

database, databases

[basis]

basis

[kind]

kinds, kind

[block]

block, blocks

[equivalence]

equivalence, equivalences

[form]

form, forms

[WSD]

WSD

[rules]

rules

[interaction]

interaction, interactions

[NLP]

NLP

[edge]

edge, edges

[resource]

Resource, resources, resource

[process]

process, processes

[idea]

idea, ideas

[baseline]

baseline, baselines

[strategy]

strategies, strategy

[understanding]

understanding

[communication]

communications, communication

[similarity]

similarity

[description]

descriptions, description

[tag]

TAGs, tag, tagging

[state]

state

[quality]

quality

[term]

term, terms

[Graph]

graphs, Graph, graph

[computation]

computation, computations

[need]

needs, need

[addition]

additions, addition

[e]

e

[implementation]

Implementation, implementation

[framework]

framework

[NIST]

NIST

[mixture]

mixture, mixtures

[homophone]

homophone

[computer]

computer, computers

[part]

part

[verb]

verb, verbs

[development]

development, developments

[noun]

noun, nouns

[English]

English

[processing]

processing

[goal]

goal, goals

[reasoning]

reasoning

[focus]

focusing, focus

[aspects]

aspects

[complexity]

complexity

[proposition]

proposition, propositions

[lexicon]

lexicon, lexicons

[CSR]

CSR

[annotation]

annotation

[precision]

precision

[world]

world

[cluster]

clustering, clusters, cluster

[BLEU]

BLEU

[pronoun]

pronoun, pronouns

[dictionary]

dictionaries, dictionary

[search]

search, searching

[issue]

issues, issue

[existing]

existing

[component]

component, components

[design]

design, designing

[effect]

effects, effect

[amount]

amount, amounts

[type]

typing, types, type

[substitution]

substitution, substitutions

[definition]

definition, definitions

[prediction]

predictions, prediction

[disposition]

disposition, dispositions

[relationship]

relationship, relationships

[fragment]

fragment, fragments

[P]

P, p

[Machine]

Machine

[evidence]

evidence

[study]

study, studies

[anaphora]

anaphora

[attention]

attention

[reduction]

reduction

[ability]

ability

[author]

authors, author

[following]

Following, following

[ATIS]

ATIS

[chart]

chart, Chart

[listener]

listener, listeners

[abstract]

abstracts, abstract

[article]

article

[properties]

properties

[construction]

construction

[unification]

unification, Unification

[summarization]

summarization

[syntax]

syntax

[comparison]

comparison

[classification]

classification

[relation]

relations, relation

[logic]

logic

[mechanism]

mechanism, mechanisms

[metaphor]

metaphors, metaphor

[class]

class, classes

[node]

node, nodes

[explanation]

explanation, explanations

[disambiguation]

disambiguation

[stemmer]

stemmer

[program]

program

[derivation]

derivations, derivation

[case]

cases, case

[transfer]

transfer, Transfer

[trigram]

trigram

[oracle]

oracle

[Category]

categories, Category, category

[area]

area, areas

[factor]

factors, factor

[segmentation]

segmentation

[newspaper]

newspapers, newspaper

[copying]

copying

[effectiveness]

effectiveness

[SPR]

SPR

[HDAG]

HDAG

[cue]

cue, cues

[memory]

memory

[CCR]

CCRs, CCR

[impact]

impact

[treatment]

treatments, treatment

[approximation]

approximation, approximations

[benefit]

benefit, benefits

[February]

February

[decoding]

decoding

[IE]

IE

[Markov]

Markov

[NE]

NE

[SI]

SI

[portion]

portion

[Web]

Web, web

[topic]

topic, Topic

[literature]

literature

[entries]

entries

[Prolog]

Prolog

[WSD]

WSD

[entity]

entities, entity

[difference]

difference

[measure]

measure

[point]

point

[hub]

hubs, hub

[solution]

solution

[intention]

intention, intentions

[HBG]

HBG

[referents]

referents

[variation]

variations, variation

[hypothesis]

hypothesis

[Chat-80]

Chat-80

[control]

control

[feedback]

feedback

[place]

place

[left]

left

[objects]

objects

[conjunction]

conjunctions, conjunction

[counterexamples]

counterexamples

[means]

means

[phrasal]

phrasal

[LFG]

LFG

[MIT]

MIT

[string]

string, strings

[actor]

actor, actors

[OCR]

OCR

[index]

indices, index

[claim]

claim, claims

[device]

device

[activity]

activity, activities

[extract]

extracts, extract

[correlation]

correlation

[conversation]

conversation

[interface]

interface, interfaces

[HPSG]

HPSG

[meeting]

meetings, meeting

[request]

request

[account]

accounts, account

[capability]

capabilities, capability

[classifiers]

classifiers

[instance]

instance

[space]

space

[prototype]

prototype

[tagger]

tagger

[view]

view

[statistics]

statistics

[name]

names, name

[scheme]

scheme

[honorifics]

honorifics, Honorifics

[n-gram]

n-gram, n-grams, N-grams

[coordinate]

coordinate

[argument]

arguments, argument

[value]

value, values

[one]

one, ones

[list]

lists, list

[content]

content, contents

[variety]

variety

[distribution]

distribution

[sample]

sample, samples

[range]

range

[L]

L

[spelling-checkers]

Spelling-checkers, spelling-checkers

[redundancy]

redundancy

[concern]

concerning, concerns, concern

[element]

element, elements

[SUN]

SUN

[paraphrasing]

paraphrasing

[Participants]

Participants, participants

[TAG]

TAG

[SENSEVAL-2]

SENSEVAL-2

[islands]

islands

[format]

format

[CFG]

CFG

[acquisition]

acquisition

[expectations]

expectations

[attempt]

attempt, attempts

[subject]

Subjects, subjects, subject

[.g]

.g

[clustering]

clustering

[predicate]

predicate, predicates

[subpredicates]

subpredicates

[progress]

progress, progressing

[agent]

agents, agent

[modification]

modifications, modification

[formulas]

formulas

[parameter]

parameters, parameter

[years]

years

[planning]

planning

[MADCOW]

MADCOW

[USR]

USR

[ensemble]

ensemble, ensembles

[requirement]

requirement, requirements

[dominance]

dominance

[protocol]

protocols, protocol

[DARPA]

DARPA

[genre]

genre

[Grammars]

Grammars

[ER]

ER

[transcription]

transcription

[WSI]

WSI

[robustness]

robustness

[LR]

LR

[QA]

QA

[Pereira]

Pereira

[LTAG]

LTAG

[assessors]

assessors

[Senseval-3]

Senseval-3

[editor]

editor

[chain]

chain, chains

[potential]

potential

[dimensions]

dimensions

[HMM]

HMM

[preference]

preference

[P-CFG]

P-CFG

[inheritance]

inheritance

[disjunctions]

disjunctions

[compound]

compound

[fact]

fact

[finding]

finding, findings

[detecting]

detecting

[criterion]

criterion, criteria

[vocabulary]

vocabulary

[interpolation]

interpolation

[entailment]

entailment

[unit]

units, unit

[identification]

identification

[IBM]

IBM

[boosting]

boosting

[style]

style, styles

[interruptions]

interruptions

[expert]

expert, experts

[levels]

levels

[query]

query, queries

[coverage]

coverage

[technology]

technology

[collocation]

collocation

[corresponding]

corresponding

[tutorial]

tutorial

[function]

function, functions

[clauses]

clauses

[principle]

principles, principle

[bootstrapping]

bootstrapping

[experience]

experience, experiences

[RAREAS]

RAREAS

[path]

path

[birds]

birds

[humans]

humans

[field]

field

[boosting]

boosting

[flexibility]

flexibility, flexibilities

[utility]

utility

[behavior]

behavior, behaviors

[mistakes]

mistakes

[morpheme]

morphemes, morpheme

[attribute]

attributes, attribute

[InfoMagnets]

InfoMagnets

[determiners]

Determiners, determiners

[board]

board

[TREC]

TREC

[hierarchy]

hierarchy, hierarchies

[image]

image

[frame]

frame

[calculus]

calculus

[Q/A]

Q/A

[tagging]

tagging

[core]

core

[link]

linking, link, Link

[cost]

costs, cost

[challenge]

challenge, challenges

[kernel]

kernel

[miscommunication]

miscommunication, miscommunications

[demonstration]

demonstration

[.5%]

.5%

[workshop]

workshops, workshop

[subtopic]

subtopic

[transition]

transition

[JAVELIN]

JAVELIN

[Japanese]

Japanese

[bigram]

bigram

[summary]

summary, summaries

[explicitation]

explicitation, Explicitation

[comma]

comma, commas

[decoder]

decoder

[network]

networks, network

[Arabic]

Arabic

[unigram]

unigram

[arc]

arcs, arc

[art]

art

[ASR]

ASR

[instantiation]

instantiation

[editing]

editing

[POURPRE]

POURPRE

[move]

move, moves

[RCL]

RCL

[surface]

surface

[entropy]

entropy

[suffix]

suffix

[nature]

nature

[generalization]

generalizations, generalization

[efficiency]

efficiency

[response]

response, responses

[lack]

lack

[learner]

learners, learner

[TDMT]

TDMT

[scratch]

scratch

[gold]

gold

[U]

U

[neighbors]

neighbors

[-speaking]

-speaking

[STTK]

STTK

[polysemy]

polysemy

[root]

root

[BIA]

BIA

[C++]

C++

[Byron's]

Byron's

[morphology]

morphology

[hypertext]

hypertext

[regard]

regard

[SPG]

SPG

[DHPL]

DHPL

[X-windows]

X-windows

[Swedes]

Swedes

[creation]

creation

[thing]

things, thing

[house]

house

[consistency]

consistency

[restoration]

restoration

[TACITUS]

TACITUS

[dags]

dags

[SUN4]

SUN4

[logistics]

logistics

[introduction]

introduction

[building]

building

[methodology]

methodology

[retrieval]

retrieval

[LCS-Marine]

LCS-Marine

[TGE]

TGE

[average]

averaging, average

[error-correction]

error-correction

[right]

right

[stage]

stages, stage

[balance]

balance

[status]

status

[USRs]

USRs

[Wall]

Wall

[Information]

Information

[informing]

informing

[industry]

industry

[NP-complete]

NP-complete

[interest]

interest

[layout]

layout

[base]

base

[stemming]

stemming, stems

[non-matches]

non-matches

[extensions]

extensions

[Korean]

Korean

[opinion]

opinion

[UCG]

UCG

[ECA]

ECA

[revision]

revision

[out-degree]

out-degree

[usefulness]

usefulness

[SYSCONJ]

SYSCONJ

[density]

density

[lexicologists]

lexicologists

[Treebank]

Treebank

[constituent]

constituent, constituents

[engine]

engine

[kanji-kana]

kanji-kana

[environment]

environment, environments

[HK]

HK

[phrasing]

phrasing

[LM]

LM

[perceptions]

perceptions

[Human-Machine]

Human-Machine

[defeasibility]

defeasibility

[PC]

PC

[HDAGs]

HDAGs

[pivot]

pivot

[[Zernik87]]

[Zernik87]

[default]

default

[metric]

metric, metrics

[TL]

TL

[Spaniards]

Spaniards

[conclusions]

conclusions

[and/or]

and/or

[newswire]

newswire

[parsing-based]

parsing-based

[Research]

Research

[porting]

porting

[section]

section

[presenter]

presenter

[system's]

system's

[Roget's]

Roget's

[generator]

generator

[four-participants]

four-participants

[extent]

extent

[assumption]

assumption

[al]

al

[scoring]

scoring

[cf]

cf

[human-machine]

human-machine

[VME]

VME

[antecedence]

antecedence

[contrary]

contrary

[specification]

specification

[IR]

IR

[belief]

belief, beliefs

[expansion]

expansion

[gaze]

gaze

[NE]

NE

[NP]

NP

[storage]

storage

[head]

head, Head

[misunderstandings]

misunderstandings

[determination]

determination

[seed]

seed, seeds

[KPSG]

KPSG

[grammaticality]

grammaticality

[TAP-XL]

TAP-XL

[TV]

TV

[failure]

failure, failures

[gap]

gap, gaps

[Examination]

Examination, examination

[insight]

insights, insight

[F-measure]

F-measure

[WER]

WER

[power]

power

[pronominalization]

pronominalization

[transformation]

transformation

[depth]

depth

[Model]

Model

[addressee]

addressee

[Senseval]

Senseval

[scale]

scale

[Fagan]

Fagan

[PageRank]

PageRank

[LIMSI]

LIMSI

[success]

success

[Analyst's]

Analyst's, analyst

[channel]

channel

[FSM]

FSM

[automaton]

automaton

[ATNs]

ATNs

[native]

native

[type-unification]

type-unification

[discussion]

discussion

[reranking]

reranking

[k-means]

k-means

[keywords]

keywords

[Collins]

Collins

[effort]

effort

[syllogism]

syllogism

[char]

char

[DoPS]

DoPS

[attendance]

attendance

[advances]

advances

[affix]

affix

[dry-run]

dry-run

[Buckley]

Buckley

[log-likelihood]

log-likelihood

[generality]

generality

[memos]

memos

[tracking]

tracking, Tracking

[Lincoln]

Lincoln

[expressiveness]

expressiveness

[magnitude]

magnitude

[refinements]

refinements

[(MT)]

(MT)

[beam-search]

beam-search

[Kyoto]

Kyoto

[coordination]

coordination

[presentation]

presentation

[independency]

independency

[word-trie]

word-trie

[laboratory]

laboratory

[certainty]

certainty

[inflection]

inflection

[Spanish]

Spanish

[Unification]

Unification

[concurrency]

concurrency

[word-sense]

word-sense, word-senses

[characterization]

characterization

[ontology]

ontology

[MRD's]

MRD's

[identifying]

identifying

[testing]

testing

[synthesis]

synthesis

[Transition]

Transition

[predicting]

predicting

[reason]

reasons, reason

[Zadeh]

Zadeh

[notions]

notions

[detection]

detection, Detection

[bag-of-words]

bag-of-words

[telephone]

telephone

[difficulty]

difficulties, difficulty

[Linde]

Linde

[vagueness]

vagueness

[hardness]

hardness

[forecaster's]

forecaster's

[architecture]

architecture

[grounding]

grounding

[correction]

correction

[limitations]

limitations

[[1]]

[1]

[quantifier]

quantifiers, quantifier

[investigation]

investigation

[programming]

programming

[Freund]

Freund

[perspective-taking]

perspective-taking

[developers]

developers

[subset]

subsets, subset

[Arguello]

Arguello

[corner]

corner

[Parsing]

Parsing

[report]

report

[publication]

publication

[Phonology]

Phonology

[reconstruction]

reconstruction

[Arabic-English]

Arabic-English

[browsing]

browsing

[/Mercer]

/Mercer

[Shieber]

Shieber

[variant]

variant

[reach]

reach

[EBMT]

EBMT

[restarting]

restarting

[observation]

observation

[passage]

passage

[typicality]

typicality

[match]

matches, match

[Czech]

Czech

[pilot]

pilot

[co-workers]

co-workers

[implications]

implications, Implications

[products]

products

[laws]

laws

[condition]

conditions, condition

[RINA]

RINA

[supply]

supply

[scenario]

scenario

[graphics]

graphics

[situation]

situation

[datasets]

datasets

[Bleu]

Bleu

[doubt]

doubt

[centering]

centering

[hardware]

hardware

[indexing]

indexing, indexings

[cohesion]

cohesion

[coherence]

coherence

[October]

October

[structure-sharing]

structure-sharing

[signals]

signals

[importance]

importance

[title]

title

[harvesting]

harvesting

[Montague]

Montague

[message-passing]

message-passing

[mentioning]

mentioning

[MAP]

MAP

[syllables]

syllables

[CDHMM]

CDHMM

[backgrounds]

backgrounds

[repeating]

repeating

[module]

module, modules

[SUMMIT]

SUMMIT

[project]

project

[enhancements]

enhancements

[MLE]

MLE

[Penn]

Penn

[speech-search]

speech-search

[LNR]

LNR

[LMs]

LMs

[Switchboard]

Switchboard

[word-trigram]

word-trigram

[benchmark]

benchmark

[KANA-KANJI]

KANA-KANJI

[MRR]

MRR

[LRE]

LRE

[voice]

voice

[Landsbergen]

Landsbergen

[sequence]

sequence

[chunk]

chunk, chunks

[annotating]

annotating

[detail]

detail

[CMU's]

CMU's

[[RCG]]

[RCG]

[discrimination]

discrimination

[Hong]

Hong

[P#P]

P#P

[mapping]

mapping

[Horn]

Horn

[lexical-cohesion]

lexical-cohesion

[boundary]

boundaries, boundary

[indexation]

indexation

[engaging]

engaging

[hastiness]

hastiness

[multiparty]

multiparty

[subparagraph]

subparagraph

[NP-]

NP-

[simplicity]

simplicity

[statement]

statement, statements

[Referring]

Referring

[Markov]

Markov

[dimensionality]

dimensionality

[[RCL]]

[RCL]

[probabilities]

probabilities

[terminal]

terminal

[Lewis]

Lewis

[rejoinder]

rejoinder, rejoinders

[consideration]

consideration, considerations

[valency]

valency

[ordering]

ordering

[decade]

decades, decade

[practice]

practice

[edit]

edit

[starting]

starting

[co-occurrence]

co-occurrence

[vector]

vector

[heuristics]

heuristics

[SmTA]

SmTA

[predicate-argument]

predicate-argument

[Korean-to-English]

Korean-to-English

[location]

location

[software]

software

[repository]

repository

[presence]

presence

[FERRET]

FERRET

[Case]

Case

[copy]

copy, copies

[Schutze's]

Schutze's

[line]

line

[recovery]

recovery

[confidence]

confidence

[Phrasal]

Phrasal

[discussing]

discussing

[Intel]

Intel

[integration]

integration

[manner]

manner

[Requestors]

Requestors

[cooccurrence]

cooccurrence

[formalism's]

formalism's

[POS]

POS

[scheduling]

scheduling

[PER)]

PER)

[.0%]

.0%

[well-formedness]

well-formedness

[.2%]

.2%

[extraction]

extraction

[light]

light

[choice]

choice

[Kumar]

Kumar

[pragmatics]

pragmatics

[California]

California

[series]

series

[SPoT]

SPoT

[session]

session

[adjoining]

adjoining, Adjoining

[material]

material

[PER]

PER

[hope]

hope

[member]

member, members

[patent]

patent

[WH-questions]

WH-questions

[direction-giving]

direction-giving

[segmenter]

segmenter

[deviations]

deviations

[LOGON]

LOGON

[people]

people

[rankings]

rankings

[operation]

operation

[demonstrator]

demonstrator

[span]

spanning, span

[principle-and-parameters]

principle-and-parameters

[extrapolation]

extrapolation

[parser/generator]

parser/generator

[scores]

scores

[TDMT]

TDMT

[uncertainty]

uncertainty

[bitexts]

bitexts

[prefix*-stem-suffix*]

prefix*-stem-suffix*

[scope]

scope

[message]

message

[age]

age

[tendency]

tendency

[likelihood]

likelihood

[SPoT]

spots, SPoT

[aid]

aid

[others]

others

[aim]

aiming, aim

[movements]

movements

[weaknesses]

weaknesses

[.62%]

.62%

[APs]

APs

[sampling]

sampling

[phone-string]

phone-string

[strength]

strength, strengths

[absence]

absence

[pre-existing]

pre-existing

[CCRs]

CCRs

[modeling]

modeling

[phone]

phone

[story-telling]

story-telling

[Navy]

Navy

[sublanguage]

sublanguages, sublanguage

[typing-algorithm]

typing-algorithm

[assignment]

assignment

[increase]

increase

[documentation]

documentation

[deal]

dealing, deal

[completion]

completion

[NTHU's]

NTHU's

[ACL]

ACL

[procedure]

procedure

[parallelism]

parallelism

[usability]

usability

[operating]

operating

[paragraph]

paragraph

[tables]

tables

[novice]

novice

[eigenvectors]

eigenvectors

[APT]

APT

[APs]

APs

[CBMT]

CBMT

[Paramax]

Paramax

[Transfer-Driven]

Transfer-Driven

[Communicator]

Communicator

[breaking-off]

breaking-off

[weather]

weather

[guide]

guide

[concatenation]

concatenation

[recall]

recall

[token]

token, tokens

[Emotions]

Emotions

[HTML]

HTML

[NOMAD]

NOMAD

[Gale]

Gale

[non-NP-antecedents]

non-NP-antecedents

[SCF]

SCF

[minor extensions of components]

minor extensions of components

[using WSD evaluation methodology]

using WSD evaluation methodology

[disambiguation scheme]

disambiguation scheme

[terms of simplicity]

terms of simplicity

[achieving similar performance to complex mixtures]

achieving similar performance to complex mixtures

[descriptors]

descriptors

[utilizing examples effectively]

utilizing examples effectively

[using a set]

using a set

[character level]

character level

[several key factors]

several key factors

[superiority of this method over methods]

superiority of this method over methods

[inflective languages such as Czech]

inflective languages such as Czech

[performance sentences]

performance sentences

[frame encoding the meaning]

frame encoding the meaning

[integral part]

integral part

[problem extracting information]

problem extracting information

[corpus analysis]

corpus analysis

[cost of hand-crafting generation systems]

cost of hand-crafting generation systems

[utterance about a task]

utterance about a task

[avoidance]

avoidance

[social relationship e .g]

social relationship e .g

[investigating the combinatorics of indexation]

investigating the combinatorics of indexation

[structures strings]

structures strings

[extraction of translation lexicons]

extraction of translation lexicons

[correcting these errors]

correcting these errors

[parsing final language]

parsing final language

[Representative samples]

Representative samples

[approach inducing a grammar]

approach inducing a grammar

[13% relative decrease]

13% relative decrease

[combination with SMT models]

combination with SMT models

[C]

C

[accurate disambiguation]

accurate disambiguation

[system support]

system support

[chart representations]

chart representations

[amount of internal variation superior]

amount of internal variation superior

[integral process in understanding]

integral process in understanding

[character recognition model]

character recognition model

[machine translation system CCLINC]

machine translation system CCLINC

[domain models with classifiers]

domain models with classifiers

[evaluation results involving automatic extraction]

evaluation results involving automatic extraction

[using processors APs for language translation]

using processors APs for language translation

[various tagging strategies]

various tagging strategies

[grammatical knowledge]

grammatical knowledge

[i efficient parsing a verb language]

i efficient parsing a verb language

[large-scale acquisition of information]

large-scale acquisition of information

[news archive]

news archive

[high cost of generation systems]

high cost of generation systems

[covering the domain]

covering the domain

[identification of 2-character names without title]

identification of 2-character names without title

[large text-understanding systems]

large text-understanding systems

[using associative processors for language translation]

using associative processors for language translation

[Arabic word segmentation system]

Arabic word segmentation system

[efficient parsing of Korean]

efficient parsing of Korean

[edges with a valid interpretation]

edges with a valid interpretation

[verb classes]

verb classes

[Markov models CDHMM]

Markov models CDHMM

[construction of models]

construction of models

[97% match accuracy]

97% match accuracy

[patent claim sentences]

patent claim sentences

[classical grammatical formalisms]

classical grammatical formalisms

[syntactic structure for speech acts]

syntactic structure for speech acts

[IDUS]

IDUS

[use including jointly conditioning on words]

use including jointly conditioning on words

[discovering parallel sentences in comparable corpora]

discovering parallel sentences in comparable corpora

[areas of research]

areas of research

[starting with a small corpus words]

starting with a small corpus words

[numeric-valued attributes such as size]

numeric-valued attributes such as size

[achieving good baseline results]

achieving good baseline results

[topmost of the edges adjacent]

topmost of the edges adjacent

[exploiting a corpus]

exploiting a corpus

[weighting of phrase translations]

weighting of phrase translations

[candidates the correct structure]

candidates the correct structure

[grammars TAG]

grammars TAG

[source of constraint]

source of constraint

[,449 word tokens]

,449 word tokens

[method describing inflection]

method describing inflection

[usually the beginning]

usually the beginning

[performance enhancing techniques]

performance enhancing techniques

[segments of tape-recorded descriptions]

segments of tape-recorded descriptions

[usual linguistic introspection]

usual linguistic introspection

[arbitrary set]

arbitrary set

[sentence condensation]

sentence condensation

[computational theory of plausible reasoning]

computational theory of plausible reasoning

[using word-trigram recognition requiring manual transcription]

using word-trigram recognition requiring manual transcription

[lexicon specification]

lexicon specification

[SLM]

SLM

[part of grammar parsing]

part of grammar parsing

[efficiency gains]

efficiency gains

[Air]

Air

[temporal relations]

temporal relations

[block identities]

block identities

[purposes a node with greater]

purposes a node with greater

[women]

women

[effect on the clusters]

effect on the clusters

[natural language in settings]

natural language in settings

[mostly women]

mostly women

[such as size]

such as size

[pints]

pints

[increase of 35%]

increase of 35%

[significant improvement for speaker adaptation]

significant improvement for speaker adaptation

[correct paraphrases meaning equivalence]

correct paraphrases meaning equivalence

[focus on considerations]

focus on considerations

[3 better indicators]

3 better indicators

[task building the editor]

task building the editor

[DCG formalism]

DCG formalism

[generation of text through transformation]

generation of text through transformation

[language learners]

language learners

[two problems numeric-valued attributes]

two problems numeric-valued attributes

[study on communications]

study on communications

[state-of-the-art Chinese word sense]

state-of-the-art Chinese word sense

[maximum a estimation]

maximum a estimation

[new algorithm for the approach]

new algorithm for the approach

[only area]

only area

[new linguistic concepts]

new linguistic concepts

[topmost of the edges]

topmost of the edges

[methods constructing word vectors LSA-based methods]

methods constructing word vectors LSA-based methods

[qualitative constraints]

qualitative constraints

[using parallel corpora a available resource]

using parallel corpora a available resource

[including conditioning on multiple consecutive words]

including conditioning on multiple consecutive words

[Markov probability of a chain]

Markov probability of a chain

[phone n-gram]

phone n-gram

[building spelling-checkers]

building spelling-checkers

[Prolog cf]

Prolog cf

[conversational cues]

conversational cues

[core modules]

core modules

[Rapid system development via acquisition]

Rapid system development via acquisition

[repository of lexical information]

repository of lexical information

[real-valued features e .g]

real-valued features e .g

[Rapid system development to domains]

Rapid system development to domains

[unsegmented texts]

unsegmented texts

[repository of information]

repository of information

[Similarity-driven Transfer System]

Similarity-driven Transfer System

[task of interpretation]

task of interpretation

[lexical approach]

lexical approach

[two contributions]

two contributions

[useful software infrastructure]

useful software infrastructure

[Overall summarization quality]

Overall summarization quality

[SRH]

SRH

[off-the-shelf classifiers]

off-the-shelf classifiers

[restrictions on the categories]

restrictions on the categories

[generating the parse]

generating the parse

[systems outputting texts]

systems outputting texts

[underlying word alignment]

underlying word alignment

[CSR Speech Recognition]

CSR Speech Recognition

[directions]

directions

[human language technology]

human language technology

[understanding of natural language]

understanding of natural language

[qualitative user evaluation]

qualitative user evaluation

[language applications]

language applications

[responding]

responding

[October '91 test set]

October '91 test set

[underspecified representation]

underspecified representation

[IDUS Document Understanding System]

IDUS Document Understanding System

[algorithms for unification]

algorithms for unification

[corpus of multiple translations]

corpus of multiple translations

[collecting statistics in a corpus]

collecting statistics in a corpus

[utterances from the target speaker]

utterances from the target speaker

[notations]

notations

[results on addressee identification in four-participants]

results on addressee identification in four-participants

[central problem of word sense]

central problem of word sense

[simple paraphrasing technique]

simple paraphrasing technique

[feature selection]

feature selection

[taking advantage of the patterns]

taking advantage of the patterns

[gathering information]

gathering information

[identifying a predicate]

identifying a predicate

[Dynamic Phrasal Lexicon]

Dynamic Phrasal Lexicon

[Recent advances]

Recent advances

[single frame encoding the meaning]

single frame encoding the meaning

[high-precision rules]

high-precision rules

[certain aspects of language learning]

certain aspects of language learning

[advantage of detailed information]

advantage of detailed information

[classical MLE reestimation algorithms]

classical MLE reestimation algorithms

[18 .3M characters]

18 .3M characters

[model using processors]

model using processors

[gain from information about meeting context]

gain from information about meeting context

[Task-based evaluation]

Task-based evaluation

[relationship ranks]

relationship ranks

[additional indices]

additional indices

[previous single tagging result]

previous single tagging result

[making exploratory corpus analysis accessible]

making exploratory corpus analysis accessible

[fewer readings]

fewer readings

[such as size perspective-taking]

such as size perspective-taking

[range of texts]

range of texts

[patterns of the parallelism]

patterns of the parallelism

[reference resolution]

reference resolution

[extracting specific information]

extracting specific information

[small size]

small size

[Diego]

Diego

[structure for speech acts]

structure for speech acts

[approach employing n-gram models]

approach employing n-gram models

[improvement in precision]

improvement in precision

[benefit from the integration]

benefit from the integration

[depending on the requirements]

depending on the requirements

[conventional word-trigram recognition]

conventional word-trigram recognition

[noun phrase reiteration]

noun phrase reiteration

[contributing matches in the sentence]

contributing matches in the sentence

[geometric model of the scene]

geometric model of the scene

[noisy portion]

noisy portion

[related areas]

related areas

[non-monotonic reasoning]

non-monotonic reasoning

[neighbour methods]

neighbour methods

[Hansards]

Hansards

[speaker leeway]

speaker leeway

[face-to-face meetings using Network classifiers]

face-to-face meetings using Network classifiers

[expensive part of grammar parsing]

expensive part of grammar parsing

[starting]

starting

[starting with a corpus 100 words]

starting with a corpus 100 words

[improvement over all]

improvement over all

[problems numeric-valued attributes]

problems numeric-valued attributes

[effects the reduction of analysis cost]

effects the reduction of analysis cost

[resulting program]

resulting program

[hand-crafted approaches]

hand-crafted approaches

[recognition accuracy]

recognition accuracy

[enumerating all possible indexings]

enumerating all possible indexings

[continuous speech recognition of a language]

continuous speech recognition of a language

[plausible interpretation]

plausible interpretation

[gathering a number of abstracts]

gathering a number of abstracts

[unification algorithms avoidance]

unification algorithms avoidance

[developing time approximations]

developing time approximations

[Computer]

Computer

[acquiring sense-tagged data]

acquiring sense-tagged data

[entering]

entering

[developing time approximations for these computations]

developing time approximations for these computations

[representing the meaning of a proposition]

representing the meaning of a proposition

[general domains]

general domains

[two problems attributes]

two problems attributes

[estimation of Markov models]

estimation of Markov models

[other NLP problems]

other NLP problems

[part-of-speech tag sequences]

part-of-speech tag sequences

[drastic generalization corresponding to arcs]

drastic generalization corresponding to arcs

[introspection]

introspection

[distribution of grammatical knowledge]

distribution of grammatical knowledge

[employing a approach]

employing a approach

[interesting ambiguities]

interesting ambiguities

[using class-dependent interpolation]

using class-dependent interpolation

[larger study of expressions]

larger study of expressions

[Robust parsing language]

Robust parsing language

[involving automatic extraction of translation lexicons]

involving automatic extraction of translation lexicons

[data objects]

data objects

[submanifold]

submanifold

[relative decrease]

relative decrease

[classical formalisms]

classical formalisms

[increase of 35% in MRR]

increase of 35% in MRR

[contributions to vocabulary]

contributions to vocabulary

[LR parser]

LR parser

[more frequently individually good predictors]

more frequently individually good predictors

[interesting combination]

interesting combination

[ideal aiming for interpretations]

ideal aiming for interpretations

[approach to restricted-domain parsing]

approach to restricted-domain parsing

[revision support]

revision support

[theory 1988/]

theory 1988/

[computer program SPRINT]

computer program SPRINT

[words with meanings]

words with meanings

[efficient parser]

efficient parser

[treatment of Montague semantics]

treatment of Montague semantics

[correct paraphrases]

correct paraphrases

[iii Rapid system development]

iii Rapid system development

[requiring less development time]

requiring less development time

[generality available]

generality available

[Using a state-of-the-art word sense]

Using a state-of-the-art word sense

[resulting edges]

resulting edges

[contrast to work]

contrast to work

[Finding the language]

Finding the language

[accuracy rate of disambiguation]

accuracy rate of disambiguation

[view for extensional reference]

view for extensional reference

[entropy word]

entropy word

[for scoring sets of concepts]

for scoring sets of concepts

[Chinese non-parallel newspaper corpora]

Chinese non-parallel newspaper corpora

[edges as in conventional treatments]

edges as in conventional treatments

[systems outputting unsegmented texts]

systems outputting unsegmented texts

[social relationship]

social relationship

[avoidance of copying of subgraphs]

avoidance of copying of subgraphs

[entries from machine dictionaries]

entries from machine dictionaries

[language generation system]

language generation system

[semantic domains]

semantic domains

[identification of Chinese names without title]

identification of Chinese names without title

[grammaticality sentences equivalence]

grammaticality sentences equivalence

[goal sounding dialog systems]

goal sounding dialog systems

[efficient Prolog cf]

efficient Prolog cf

[process in understanding]

process in understanding

[m-th]

m-th

[confidence measures]

confidence measures

[proper treatment]

proper treatment

[third]

third

[small corpus 100 words]

small corpus 100 words

[utterances from five sites]

utterances from five sites

[approach to Q/A]

approach to Q/A

[CDD]

CDD

[move in light]

move in light

[many properties]

many properties

[efficient parsing language with case markers]

efficient parsing language with case markers

[employing n-gram models]

employing n-gram models

[state Arabic stemmer]

state Arabic stemmer

[reliable basis for the acquisition]

reliable basis for the acquisition

[planning COMPLEX a lexicon]

planning COMPLEX a lexicon

[word-senses of words with meanings]

word-senses of words with meanings

[Cooccurrence]

Cooccurrence

[improvement in paraphrase classification accuracy]

improvement in paraphrase classification accuracy

[unigram model]

unigram model

[speech tagging]

speech tagging

[translation examples]

translation examples

[SemCor]

SemCor

[corpus exploration tool]

corpus exploration tool

[Hierarchical Graph Kernel]

Hierarchical Graph Kernel

[Several experiments corresponding to domains]

Several experiments corresponding to domains

[choice for the parser]

choice for the parser

[honorifics words]

honorifics words

[large list]

large list

[applying statistical models]

applying statistical models

[employing a traditional approach]

employing a traditional approach

[information related to a scenario]

information related to a scenario

[proprietary Arabic stemmer]

proprietary Arabic stemmer

[speech recognition applications]

speech recognition applications

[applying a set]

applying a set

[definition a concept]

definition a concept

[posing problems]

posing problems

[corresponding to arcs in a chart]

corresponding to arcs in a chart

[making corpus analysis accessible to researchers]

making corpus analysis accessible to researchers

[retrieving examples similar]

retrieving examples similar

[key factors behind this surprising finding]

key factors behind this surprising finding

[addressee identification in four-participants]

addressee identification in four-participants

[linguistic parser/generator for LFG]

linguistic parser/generator for LFG

[mixture observation densities]

mixture observation densities

[word-trigram recognition requiring manual transcription]

word-trigram recognition requiring manual transcription

[broad-coverage principle-based parser]

broad-coverage principle-based parser

[representation the Dynamic Phrasal Lexicon]

representation the Dynamic Phrasal Lexicon

[Transfer System]

Transfer System

[node with in-degree greater]

node with in-degree greater

[representing the meaning]

representing the meaning

[standard floppy]

standard floppy

[Web-based writing]

Web-based writing

[also in the related areas]

also in the related areas

[University]

University

[shallow linguistic features]

shallow linguistic features

[levels of representation]

levels of representation

[adequate explanation of discourse phenomena]

adequate explanation of discourse phenomena

[he/she/man/woman]

he/she/man/woman

[knowledgebases]

knowledgebases

[workability]

workability

[gain from information]

gain from information

[priors in conditional models]

priors in conditional models

[decoder implementations]

decoder implementations

[representation USR]

representation USR

[predicting boundaries]

predicting boundaries

[phrase translations from alignments]

phrase translations from alignments

[yielding a treatment]

yielding a treatment

[role of information]

role of information

[consistency problem]

consistency problem

[attractive properties]

attractive properties

[i Robust parsing]

i Robust parsing

[only 12 training speakers]

only 12 training speakers

[mutually equivalent readings]

mutually equivalent readings

[evaluation results]

evaluation results

[subsets of database]

subsets of database

[copying of subgraphs]

copying of subgraphs

[little corpus]

little corpus

[adverbs]

adverbs

[Natural Language Processing NLP]

Natural Language Processing NLP

[existing spelling-checkers]

existing spelling-checkers

[constraints as levels]

constraints as levels

[method discovering parallel sentences]

method discovering parallel sentences

[kind of Role System]

kind of Role System

[social ranks]

social ranks

[two grammatical formalisms]

two grammatical formalisms

[relationship ranks of the referents]

relationship ranks of the referents

[newspaper articles on missiles]

newspaper articles on missiles

[MLE reestimation algorithms]

MLE reestimation algorithms

[evaluating answers to definition questions]

evaluating answers to definition questions

[concepts of the linguistic theory]

concepts of the linguistic theory

[intervals]

intervals

[WSI algorithm]

WSI algorithm

[sparsity]

sparsity

[using semantic categories]

using semantic categories

[Markov Model]

Markov Model

[simple unsupervised technique]

simple unsupervised technique

[amount of ambiguity]

amount of ambiguity

[log(]

log(

[automating aspects of language learning]

automating aspects of language learning

[computers with graphics displays]

computers with graphics displays

[building corpus from HTML documents]

building corpus from HTML documents

[part of a larger study]

part of a larger study

[only 40 utterances]

only 40 utterances

[capabilities in dialogue systems]

capabilities in dialogue systems

[acquiring stems]

acquiring stems

[scope information]

scope information

[assessing developments]

assessing developments

[equivalent readings]

equivalent readings

[human understanding of language]

human understanding of language

[outputs systems entity annotations]

outputs systems entity annotations

[.5% word]

.5% word

[copying of unmodified subgraphs]

copying of unmodified subgraphs

[parsing accuracy rate]

parsing accuracy rate

[polynomial time solution]

polynomial time solution

[specific domain]

specific domain

[two lists]

two lists

[application of the techniques]

application of the techniques

[advantage of this approach]

advantage of this approach

[.0% improvement over system]

.0% improvement over system

[log-linear models]

log-linear models

[dialogue data]

dialogue data

[visualization method]

visualization method

[work harvesting bitexts]

work harvesting bitexts

[interaction with data sources]

interaction with data sources

[three techniques]

three techniques

[case-based reasoning]

case-based reasoning

[Entity NE]

Entity NE

[chart Parsing instantiation]

chart Parsing instantiation

[dynamic challenge]

dynamic challenge

[effects the reduction]

effects the reduction

[360K floppy]

360K floppy

[paradigm]

paradigm

[parallel sentences in corpora]

parallel sentences in corpora

[great deal of language texts]

great deal of language texts

[12,000 utterances of spontaneous speech]

12,000 utterances of spontaneous speech

[coverage grammar]

coverage grammar

[unrestricted texts]

unrestricted texts

[Prolog subset]

Prolog subset

[importantly texts]

importantly texts

[TDMT on APs]

TDMT on APs

[more questions]

more questions

[statistical machine translation method]

statistical machine translation method

[using processors]

using processors

[way automatically identifying argument structures]

way automatically identifying argument structures

[reference word]

reference word

[.e]

.e

[access to a news archive]

access to a news archive

[domain independent models]

domain independent models

[12,000 utterances from sites]

12,000 utterances from sites

[Using utterances from the target speaker]

Using utterances from the target speaker

[neat relation to grammar]

neat relation to grammar

[sample output]

sample output

[novel method adding linguistic annotation]

novel method adding linguistic annotation

[log-linear interpolation]

log-linear interpolation

[Syntactic analysis of the coordinate sentences]

Syntactic analysis of the coordinate sentences

[Robust parsing of Korean a verb]

Robust parsing of Korean a verb

[model of language]

model of language

[same nodes]

same nodes

[new measure with judgment]

new measure with judgment

[techniques for syntactic analysis]

techniques for syntactic analysis

[data structure for machine translation]

data structure for machine translation

[flexible planning-based architecture]

flexible planning-based architecture

[performing cluster number estimation]

performing cluster number estimation

[interpretation of natural language]

interpretation of natural language

[algorithm acquiring new stems]

algorithm acquiring new stems

[shortcomings]

shortcomings

[standard term]

standard term

[parameter smoothing]

parameter smoothing

[little corpus of words]

little corpus of words

[human-computer interaction]

human-computer interaction

[method building polarity-tagged corpus]

method building polarity-tagged corpus

[using a fake symbol]

using a fake symbol

[corpus of 2 SRHs]

corpus of 2 SRHs

[Hierarchical Graph]

Hierarchical Graph

[factor of speed-up]

factor of speed-up

[providing useful insights]

providing useful insights

[various kinds of misunderstandings]

various kinds of misunderstandings

[existing approaches]

existing approaches

[most men]

most men

[words ellipsis]

words ellipsis

[applying standard MT evaluation methods]

applying standard MT evaluation methods

[better results]

better results

[incorporating the control structure]

incorporating the control structure

[fake non-terminal symbol]

fake non-terminal symbol

[NLP tasks]

NLP tasks

[language in settings]

language in settings

[use of semantics]

use of semantics

[ellipses]

ellipses

[negligible runtime]

negligible runtime

[covering the domain of HK laws]

covering the domain of HK laws

[treatment for sentence analyses]

treatment for sentence analyses

[strategies utilizing mechanisms]

strategies utilizing mechanisms

[learning methodology applicable in domains]

learning methodology applicable in domains

[5K]

5K

[means learning of phrase translations]

means learning of phrase translations

[problem of word sense WSD]

problem of word sense WSD

[avoidance of copying]

avoidance of copying

[information from a inflow of data]

information from a inflow of data

[grammaticality equivalence]

grammaticality equivalence

[speed]

speed

[alignment techniques]

alignment techniques

[resolving this ambiguity]

resolving this ambiguity

[internal variation]

internal variation

[representative SMT model]

representative SMT model

[channel framework]

channel framework

[Q/A applications]

Q/A applications

[lack of data]

lack of data

[gathering a number from the Web]

gathering a number from the Web

[algorithm for the approach]

algorithm for the approach

[expressiveness of the two formalisms]

expressiveness of the two formalisms

[two strategies]

two strategies

[efficient parsing]

efficient parsing

[loss]

loss

[language input]

language input

[example from an fragment]

example from an fragment

[consisting of 126 sentences]

consisting of 126 sentences

[State Model FSM]

State Model FSM

[independent event-based summarization approaches]

independent event-based summarization approaches

[experience in the MUC-3 evaluation]

experience in the MUC-3 evaluation

[detection of those activities]

detection of those activities

[list of possible sentence plans]

list of possible sentence plans

[Representative samples from an language definition]

Representative samples from an language definition

[considerable importance]

considerable importance

[NE .g]

NE .g

[relationship social ranks]

relationship social ranks

[translation into an RCG]

translation into an RCG

[other syntax formalisms]

other syntax formalisms

[formal computation]

formal computation

[underspecified semantic representation USR]

underspecified semantic representation USR

[foundation]

foundation

[probabilistic context-free grammars]

probabilistic context-free grammars

[many lexical-syntactic forms]

many lexical-syntactic forms

[simple technique]

simple technique

[use of parser evaluation methods]

use of parser evaluation methods

[contributions continuous speech recognition]

contributions continuous speech recognition

[lexical variation in a set]

lexical variation in a set

[96% paraphrases meaning equivalence]

96% paraphrases meaning equivalence

[1 an set]

1 an set

[potential increase of 35%]

potential increase of 35%

[extending the speech/language integration architecture]

extending the speech/language integration architecture

[features from the contexts]

features from the contexts

[remote meteorological events]

remote meteorological events

[forcing a interpretation]

forcing a interpretation

[two problems attributes such as size]

two problems attributes such as size

[numeric-valued attributes]

numeric-valued attributes

[existence]

existence

[parsing a verb final language]

parsing a verb final language

[iii system development to new domains]

iii system development to new domains

[pharmaceutical news archive]

pharmaceutical news archive

[broad use including conditioning on words]

broad use including conditioning on words

[ground in interaction]

ground in interaction

[DCG]

DCG

[model of the process]

model of the process

[misunderstandings between speaker]

misunderstandings between speaker

[suffix data structure]

suffix data structure

[Document Understanding System]

Document Understanding System

[insertion]

insertion

[use in a multi-site evaluation]

use in a multi-site evaluation

[processors APs]

processors APs

[configuration]

configuration

[exploratory corpus analysis accessible]

exploratory corpus analysis accessible

[language understanding with speech recognition]

language understanding with speech recognition

[relations between segments]

relations between segments

[using a speech]

using a speech

[also a generalization corresponding to arcs]

also a generalization corresponding to arcs

[relatively simple means heuristic learning]

relatively simple means heuristic learning

[issue recently]

issue recently

[effective use]

effective use

[chemical warfare]

chemical warfare

[theory of human reasoning]

theory of human reasoning

[first empirical test]

first empirical test

[DFA]

DFA

[corpora a available resource]

corpora a available resource

[Robust parsing language with case markers]

Robust parsing language with case markers

[coordinate structure]

coordinate structure

[applying mutual information]

applying mutual information

[literal meaning]

literal meaning

[relatively free word]

relatively free word

[multiple English translations]

multiple English translations

[.5% agreement]

.5% agreement

[claims to the contrary]

claims to the contrary

[chart-based phrase structure parsing]

chart-based phrase structure parsing

[analysis cost]

analysis cost

[communities]

communities

[version]

version

[stop]

stop

[dominant part]

dominant part

[MT evaluation methods]

MT evaluation methods

[meaning of input]

meaning of input

[scoring speech recognition hypotheses in terms]

scoring speech recognition hypotheses in terms

[interactive corpus exploration tool]

interactive corpus exploration tool

[concepts of the underlying theory]

concepts of the underlying theory

[Phrasal Lexicon [Zernik88]]

Phrasal Lexicon [Zernik88]

[most men like young women]

most men like young women

[feature selection methods]

feature selection methods

[problems including unlimited vocabulary]

problems including unlimited vocabulary

[answers in corpora]

answers in corpora

[BU]

BU

[clustering verbs]

clustering verbs

[extensional equivalence]

extensional equivalence

[certain kinds]

certain kinds

[syntax-based constraint]

syntax-based constraint

[forming an utterance]

forming an utterance

[Basque]

Basque

[query optimisation]

query optimisation

[condition for this test suite]

condition for this test suite

[computational system speech recognition]

computational system speech recognition

[reduction of analysis cost]

reduction of analysis cost

[contributions to vocabulary speech recognition]

contributions to vocabulary speech recognition

[standard WSD evaluation methodology]

standard WSD evaluation methodology

[assessment]

assessment

[plausible reasoning]

plausible reasoning

[standard outputs from information extraction]

standard outputs from information extraction

[dealing with complexity]

dealing with complexity

[meanings ambiguity]

meanings ambiguity

[using complicated knowledgebases]

using complicated knowledgebases

[means learning]

means learning

[Marker]

Marker

[mixture of these techniques]

mixture of these techniques

[syntax-based machine translation system]

syntax-based machine translation system

[biasing parsing toward meanings]

biasing parsing toward meanings

[linguistic parser/generator]

linguistic parser/generator

[i Robust parsing language]

i Robust parsing language

[i parsing of Korean a verb]

i parsing of Korean a verb

[relation to categorial grammar]

relation to categorial grammar

[useful features]

useful features

[computer text generation system]

computer text generation system

[Hierarchical Phrasal Lexicon DHPL [Zernik88]]

Hierarchical Phrasal Lexicon DHPL [Zernik88]

[Research Methods course]

Research Methods course

[achieving robustness]

achieving robustness

[five sites]

five sites

[spots in language processing]

spots in language processing

[parallel sentences in comparable corpora]

parallel sentences in comparable corpora

[disambiguation systems one]

disambiguation systems one

[unknown word features]

unknown word features

[three methods constructing word vectors]

three methods constructing word vectors

[still satisfying communication needs]

still satisfying communication needs

[equivalence classification]

equivalence classification

[language models LMs]

language models LMs

[constraint for word alignment]

constraint for word alignment

[SENSEVAL-2 lexical sample task]

SENSEVAL-2 lexical sample task

[adjacency]

adjacency

[subset .0%]

subset .0%

[possible variations]

possible variations

[numerical constraints]

numerical constraints

[old Resource Management RM corpus]

old Resource Management RM corpus

[word unsegmented corpus]

word unsegmented corpus

[most important task]

most important task

[recent debate]

recent debate

[linking of entries]

linking of entries

[parser for a language]

parser for a language

[method for speech understanding]

method for speech understanding

[identifying hubs]

identifying hubs

[research on new problems]

research on new problems

[NE he/she/man/woman]

NE he/she/man/woman

[material the Hansards]

material the Hansards

[system combination for WSD]

system combination for WSD

[document descriptors]

document descriptors

[different lexical-syntactic forms]

different lexical-syntactic forms

['synchronous' way]

'synchronous' way

[average precision]

average precision

[new view for reference]

new view for reference

[three words from word-level alignment models]

three words from word-level alignment models

[interpretation of ellipses]

interpretation of ellipses

[event-based mutli-document summarization approaches]

event-based mutli-document summarization approaches

[non-parallel newspaper corpora]

non-parallel newspaper corpora

[Syntax-based machine translation MT]

Syntax-based machine translation MT, syntax-based machine translation MT

[data relevant]

data relevant

[C code]

C code

[combining language models LMs]

combining language models LMs

[chart Parsing instantiation of clauses]

chart Parsing instantiation of clauses

[saturation]

saturation

[exploitation]

exploitation

[robust parsing method]

robust parsing method

[visualizing]

visualizing

[superiority over methods]

superiority over methods

[integrating language understanding]

integrating language understanding

[in-domain corpus]

in-domain corpus

[quality translation]

quality translation

[dialog model]

dialog model

[corresponding with the best confidence]

corresponding with the best confidence

[method for analysis]

method for analysis

[particular scenario]

particular scenario

[word-semantic information]

word-semantic information

[tree-to-tree translation]

tree-to-tree translation

[centering on the construction]

centering on the construction

[performance of models]

performance of models

[abstracting]

abstracting

[move structures]

move structures

[representation of documents]

representation of documents, representations of documents

[articulation]

articulation

[appealing alternative—in terms]

appealing alternative—in terms

[use of substitutions]

use of substitutions

[monitoring of lack]

monitoring of lack

[using features of the tree]

using features of the tree

[Understanding]

Understanding

[efficient principle-based parser for English]

efficient principle-based parser for English

[test of an speculative claim]

test of an speculative claim

[abstraction]

abstraction

[subject of debate]

subject of debate

[method producing phrases]

method producing phrases

[own hierarchy]

own hierarchy

[High quality translation via word sense]

High quality translation via word sense

[i Robust parsing of Korean]

i Robust parsing of Korean

[numerous aspects]

numerous aspects

[saturation of the structure]

saturation of the structure

[bearing on commonsense reasoning]

bearing on commonsense reasoning

[even in a way]

even in a way

[Role System]

Role System

[past years]

past years

[answering agents searching for answers]

answering agents searching for answers

[identifying predicate argument structures]

identifying predicate argument structures

[effective CFG filter]

effective CFG filter

[containing 28 word tokens]

containing 28 word tokens

[Acyclic Graph HDAG]

Acyclic Graph HDAG

[large unsegmented corpus]

large unsegmented corpus

[monitoring]

monitoring

[therefore a choice]

therefore a choice

[eye]

eye

[progressing from generation of text]

progressing from generation of text

[using processors for language translation]

using processors for language translation

[machine translation evaluation]

machine translation evaluation

[powerful computers with graphics displays]

powerful computers with graphics displays

[longest span]

longest span

[hand-crafting generation systems]

hand-crafting generation systems

[SA]

SA

[two outputs systems entity annotations]

two outputs systems entity annotations

[SL]

SL

[approach to ranking problems]

approach to ranking problems

[backbone]

backbone

[model’s]

model’s

[F-measure a relative decrease]

F-measure a relative decrease

[cohesion-based approach]

cohesion-based approach

[interpretation for values]

interpretation for values

[large Arabic corpus]

large Arabic corpus

[Two hardness results]

Two hardness results

[use in MT]

use in MT

[new technology of computers]

new technology of computers

[models of WH-questions]

models of WH-questions

[member of the set]

member of the set

[Robust parsing of Korean]

Robust parsing of Korean

[special problems for readers]

special problems for readers

[.8% improvement]

.8% improvement

[indices such as the dominance distribution]

indices such as the dominance distribution

[texts in the domain]

texts in the domain

[performance of the word-sense]

performance of the word-sense

[language model of an corpus]

language model of an corpus

[sense per collocation observation]

sense per collocation observation

[limited domain semantics]

limited domain semantics

[pairs of grammars working]

pairs of grammars working

[particular interpretation]

particular interpretation

[results on addressee identification face-to-face meetings]

results on addressee identification face-to-face meetings

[increasing the dialogue duration]

increasing the dialogue duration

[coordinate structure analysis model]

coordinate structure analysis model

[approach to parsing]

approach to parsing

[lexical variation]

lexical variation

[pronoun in sentences]

pronoun in sentences

[systems entity annotations]

systems entity annotations

[language technology]

language technology

[meta-inference types]

meta-inference types

[[Zernik88]]

[Zernik88]

[inflow]

inflow

[system with mobile agents]

system with mobile agents

[statistical POS tagger]

statistical POS tagger

[gold standard]

gold standard

[Phonology Structure Grammar]

Phonology Structure Grammar

[achieving performance to mixtures]

achieving performance to mixtures

[speech recognition hypotheses SRH]

speech recognition hypotheses SRH

[WI]

WI

[fine-grained modeling]

fine-grained modeling

[version of grammars]

version of grammars

[condensation]

condensation

[anaphora resolution module]

anaphora resolution module

[employing a phonological approach]

employing a phonological approach

[mutli-document summarization]

mutli-document summarization

[representation the Hierarchical Phrasal Lexicon]

representation the Hierarchical Phrasal Lexicon

[Statistical Machine Translation]

Statistical Machine Translation

[news]

news

[Speech Recognition technology]

Speech Recognition technology

[existing probabilistic parsing models]

existing probabilistic parsing models

[Chomsky's program]

Chomsky's program

[larger study]

larger study

[Senseval-3 lexical sample task]

Senseval-3 lexical sample task

[Learning Model]

Learning Model

[communications between APs]

communications between APs

[according to character bigrams]

according to character bigrams

[practical approach]

practical approach

[simple means]

simple means

[effect using an corpus]

effect using an corpus

[many formalisms]

many formalisms

[examples most similar]

examples most similar

[plan]

plan

[agenda-based scheduling parser]

agenda-based scheduling parser

[weather data]

weather data

[answer candidates]

answer candidates

[language learning process]

language learning process

[responses adaptive]

responses adaptive

[portable understanding system]

portable understanding system

[corpus data]

corpus data

[commonly available resource]

commonly available resource

[superset]

superset

[.24% accuracy]

.24% accuracy

[gain]

gain

[technology of computers]

technology of computers

[adopting a paraphrasing technique]

adopting a paraphrasing technique

[context-dependent modelling]

context-dependent modelling

[FlexP]

FlexP

[candidates the structure]

candidates the structure

[introduction to machine translation]

introduction to machine translation

[Tree Grammars]

Tree Grammars

[including inherent limitations of MT architectures]

including inherent limitations of MT architectures

[generative model of natural language]

generative model of natural language

[annotation performance]

annotation performance

[different strategies utilizing mechanisms]

different strategies utilizing mechanisms

[distinction]

distinction

[new view]

new view

[text-understanding systems]

text-understanding systems

[most young men]

most young men

[Robust parsing a verb language]

Robust parsing a verb language

[meeting situation]

meeting situation

[Two styles performing inference in networks]

Two styles performing inference in networks

[common attribute sequences]

common attribute sequences

[interpretation from a chunk]

interpretation from a chunk

[improvement over word alignment techniques]

improvement over word alignment techniques

[F-measure a 13% decrease]

F-measure a 13% decrease

[quantifier most in the proposition]

quantifier most in the proposition

[two standard outputs from information extraction]

two standard outputs from information extraction

[holding multiple candidates]

holding multiple candidates

[efficient Prolog]

efficient Prolog

[knowledge base system]

knowledge base system

[use including conditioning on multiple words]

use including conditioning on multiple words

[using a speech from many speakers]

using a speech from many speakers

[research on problems including vocabulary]

research on problems including vocabulary

[dimensions of user models]

dimensions of user models

[contributing]

contributing

[possible candidates]

possible candidates

[two word-sense disambiguation systems one]

two word-sense disambiguation systems one

[explication]

explication

[special type of automaton]

special type of automaton

[applying models to data]

applying models to data

[valuable methods]

valuable methods

[frequently predictors of document descriptors]

frequently predictors of document descriptors

[applying MT evaluation methods]

applying MT evaluation methods

[State Model]

State Model

[Stabler's]

Stabler's

[categories on the terminal]

categories on the terminal

[cluster number estimation]

cluster number estimation

[simple means learning]

simple means learning

[repository of information for use]

repository of information for use

[fulfillment]

fulfillment

[subjective judgments]

subjective judgments

[amount of work]

amount of work

[Using utterances for adaptation]

Using utterances for adaptation

[predicting the right translation]

predicting the right translation

[efficient parsing a verb]

efficient parsing a verb

[special problems]

special problems

[requirements of language translation]

requirements of language translation

[monolingual material]

monolingual material

[efficient Prolog query optimisation]

efficient Prolog query optimisation

[employing a traditional generative approach]

employing a traditional generative approach

[tag sequences]

tag sequences

[37% reduction]

37% reduction

[estimation of Markov models CDHMM]

estimation of Markov models CDHMM

[ground in human-computer interaction]

ground in human-computer interaction

[idea of this method]

idea of this method

[discovering parallel sentences in corpora]

discovering parallel sentences in corpora

[parsing of Korean a verb language]

parsing of Korean a verb language

[High quality translation]

High quality translation

[small 10K sentences parallel corpus]

small 10K sentences parallel corpus

[MUC-3 evaluation]

MUC-3 evaluation

[using ASR output]

using ASR output

[redundancy elimination]

redundancy elimination

[dimensions of user models skill level]

dimensions of user models skill level

[features a language model score]

features a language model score

[areas of Human-Machine Communication]

areas of Human-Machine Communication

[difficult problems]

difficult problems

[database query]

database query

[also a drastic generalization roughly corresponding]

also a drastic generalization roughly corresponding

[Graph Kernel]

Graph Kernel

[noun group]

noun group

[large knowledgebases]

large knowledgebases

[actual descriptions]

actual descriptions

[fonts]

fonts

[transducer]

transducer

[applying a set of extraction]

applying a set of extraction

[accuracy in language identification]

accuracy in language identification

[users of tool]

users of tool

[vital requirements]

vital requirements

[representational format]

representational format

[application of Unification Grammar]

application of Unification Grammar

[evaluating the summarization quality]

evaluating the summarization quality

[similarity between words]

similarity between words

[piece]

piece

[role in syntax]

role in syntax

[.0% improvement]

.0% improvement

[representation of a scope ambiguity]

representation of a scope ambiguity

[question answering capability on text]

question answering capability on text

[describing the layout]

describing the layout

[language understanding technology]

language understanding technology

[involving extraction from text]

involving extraction from text

[relationship e .g ranks]

relationship e .g ranks

[skill]

skill

[responses adaptive to users]

responses adaptive to users

[using a a corpus]

using a a corpus

[instantiation scheme]

instantiation scheme

[new part-of-speech tagger use]

new part-of-speech tagger use

[system development via acquisition of grammars]

system development via acquisition of grammars

[guidance for novice users]

guidance for novice users

[Web-based academic writing]

Web-based academic writing

[improvement for speaker adaptation SA]

improvement for speaker adaptation SA

[further reduction]

further reduction

[several factors including inherent limitations]

several factors including inherent limitations

[finding data]

finding data

[Begin/After]

Begin/After

[promising approach to writing]

promising approach to writing

[language parsing NLP]

language parsing NLP

[evaluating a SMT model]

evaluating a SMT model

[precise description]

precise description

[certain classes]

certain classes

[5 years]

5 years

[grammaticality correct sentences equivalence]

grammaticality correct sentences equivalence

[construction of a corpus resource]

construction of a corpus resource

[Chinese-English texts]

Chinese-English texts

[New experimental results on all applications]

New experimental results on all applications

[possible structures]

possible structures

[meaning of a sentence]

meaning of a sentence

[contrast]

contrast

[sense of as much]

sense of as much

[sentence co-occurrences]

sentence co-occurrences

[28 word tokens]

28 word tokens

[tagging result]

tagging result

[including Language Processing]

including Language Processing

[occurrences]

occurrences

[methods .e LSA-based methods]

methods .e LSA-based methods

[reflecting the relationship social ranks]

reflecting the relationship social ranks

[scheme's]

scheme's

[predictors of document descriptors]

predictors of document descriptors

[text retrieval]

text retrieval

[according to simple character bigrams]

according to simple character bigrams

[arbiter-based combination]

arbiter-based combination

[person's knowledge]

person's knowledge

[discovering parallel sentences]

discovering parallel sentences

[accomplishments]

accomplishments

[LM]

LM

[evaluation using information retrieval]

evaluation using information retrieval

[different algorithms]

different algorithms

[good guidance]

good guidance

[adding annotation]

adding annotation

[personal computers with graphics displays]

personal computers with graphics displays

[output of OCR systems]

output of OCR systems

[extending activity]

extending activity

[planning COMPLEX]

planning COMPLEX

[MT output]

MT output

[amount of lexical variation superior]

amount of lexical variation superior

[domain acoustic models with off-the-shelf classifiers]

domain acoustic models with off-the-shelf classifiers

[inducing a grammar]

inducing a grammar

[Generation of Expressions]

Generation of Expressions

[syntactic construction]

syntactic construction

[partial instantiation]

partial instantiation

[Phonology Structure Grammar KPSG]

Phonology Structure Grammar KPSG

[superiority of this method]

superiority of this method

[such as editing terms]

such as editing terms

[method of inference]

method of inference

[importance to Machine Translation SMT]

importance to Machine Translation SMT

[frequent omissions]

frequent omissions

[efficient broad-coverage principle-based parser]

efficient broad-coverage principle-based parser

[great benefit]

great benefit

[broad use of features including conditioning]

broad use of features including conditioning

[context-free grammars working]

context-free grammars working

[contributions speech recognition]

contributions speech recognition

[using Bayesian Network classifiers]

using Bayesian Network classifiers

[many different lexical-syntactic forms]

many different lexical-syntactic forms

[automatic word sense induction]

automatic word sense induction

[generative probabilistic character recognition model]

generative probabilistic character recognition model

[also a generalization corresponding]

also a generalization corresponding

[candidates the syntactic structure]

candidates the syntactic structure

[Senseval series]

Senseval series

[TDMT the part]

TDMT the part

[suppressing the quantifier]

suppressing the quantifier

[2-character names without title]

2-character names without title

[Slavonic languages]

Slavonic languages

[representation the Phrasal Lexicon [Zernik88]]

representation the Phrasal Lexicon [Zernik88]

[adaptation]

adaptation

[training learners]

training learners

[model using processors APs]

model using processors APs

[system users]

system users

[semantically classifying SCF data]

semantically classifying SCF data

[relative decrease in F-measure error]

relative decrease in F-measure error

[least 96% paraphrases meaning equivalence]

least 96% paraphrases meaning equivalence

[judicious mixture]

judicious mixture

[Korean newspaper articles]

Korean newspaper articles

[tree-based models]

tree-based models

[correctness of answers]

correctness of answers

[corpus analysis accessible]

corpus analysis accessible

[presence of words]

presence of words

[new word-sense disambiguation systems one]

new word-sense disambiguation systems one

[encoding the meaning of the sentence]

encoding the meaning of the sentence

[computational model]

computational model

[substantial improvement over all]

substantial improvement over all

[better translation quality]

better translation quality

[RM]

RM

[flexible architecture with a variety]

flexible architecture with a variety

[abduction of relations between segments]

abduction of relations between segments

[repository of data objects]

repository of data objects

[parsing a verb language]

parsing a verb language

[comparison of CFG filtering techniques]

comparison of CFG filtering techniques

[lack of feedback]

lack of feedback

[generative optical character recognition model]

generative optical character recognition model

[Stabler's formalization]

Stabler's formalization

[languages with little inflection]

languages with little inflection

[new disambiguation scheme]

new disambiguation scheme

[Communication]

Communication

[volume from the Web]

volume from the Web

[Phrasal Lexicon]

Phrasal Lexicon

[collection of 12,000 utterances]

collection of 12,000 utterances

[Other clues]

Other clues

[MT architectures]

MT architectures

[analyzing English sentences]

analyzing English sentences

[one language]

one language

[translation equivalents]

translation equivalents

[distinct domains tutorial dialogue]

distinct domains tutorial dialogue

[Graph Kernel for language data]

Graph Kernel for language data

[Several experiments corresponding]

Several experiments corresponding

[equivalence of intensional concepts]

equivalence of intensional concepts

[example from an English-Spanish fragment]

example from an English-Spanish fragment

[negative impact]

negative impact

[social relationship e .g ranks]

social relationship e .g ranks

[grammaticality equivalence in meaning]

grammaticality equivalence in meaning

[information from a inflow]

information from a inflow

[searching for answers]

searching for answers

[other methods]

other methods

[achieving performance to more complex mixtures]

achieving performance to more complex mixtures

[benefit to language pairs]

benefit to language pairs

[semantics of natural language applications]

semantics of natural language applications

[four effects]

four effects

[problems a attributes such as size]

problems a attributes such as size

[necessary conditions for the use]

necessary conditions for the use

[organizational strategies]

organizational strategies

[new disambiguation systems]

new disambiguation systems

[theory about these rules]

theory about these rules

[multiple words]

multiple words

[Monolingual text]

Monolingual text

[Deictic reference]

Deictic reference

[candidates for the understanding result]

candidates for the understanding result

[user studies]

user studies

[anaphora resolution module of Lucy]

anaphora resolution module of Lucy

[extracting information from texts]

extracting information from texts

[readings]

readings

[domain semantics]

domain semantics

[potentially large list]

potentially large list

[correlation between the parse-based evaluation]

correlation between the parse-based evaluation

[equivalence of concepts]

equivalence of concepts

[reducing the amount of ambiguity]

reducing the amount of ambiguity

[strategies utilizing knowledge-based mechanisms]

strategies utilizing knowledge-based mechanisms

[constructing a phonological system]

constructing a phonological system

[using a LM]

using a LM

[point of view]

point of view

[recall of 49%]

recall of 49%

[probabilistic dependency insertion grammar]

probabilistic dependency insertion grammar

[.3M Chinese characters]

.3M Chinese characters

[factors behind this finding]

factors behind this finding

[phases]

phases

[principled techniques for analysis]

principled techniques for analysis

[University at San Diego]

University at San Diego

[training two learners]

training two learners

[new word-sense disambiguation systems]

new word-sense disambiguation systems

[grammaticality at least 99% sentences]

grammaticality at least 99% sentences

[scenic descriptions]

scenic descriptions

[ranking problems]

ranking problems

[detailed information]

detailed information

[recent work harvesting English-Chinese bitexts]

recent work harvesting English-Chinese bitexts

[robust interactive method]

robust interactive method

[system development via knowledge-based acquisition]

system development via knowledge-based acquisition

[finding more data relevant]

finding more data relevant

[scoring speech recognition hypotheses SRH]

scoring speech recognition hypotheses SRH

[exploitation of domain semantics]

exploitation of domain semantics

[kinds of knowledge]

kinds of knowledge

[knowledge about own knowledge]

knowledge about own knowledge

[parsing algorithms]

parsing algorithms

[basis for a useful tool]

basis for a useful tool

[results of a qualitative user evaluation]

results of a qualitative user evaluation

[speech/language integration]

speech/language integration

[Roget's Thesaurus]

Roget's Thesaurus

[explicit information from dictionaries]

explicit information from dictionaries

[results of a preliminary user evaluation]

results of a preliminary user evaluation

[corpus 100 ,000 words]

corpus 100 ,000 words

[application to the framework]

application to the framework

[extracting specific information from texts]

extracting specific information from texts

[maximum-entropy model for output selection]

maximum-entropy model for output selection

[novel method building corpus]

novel method building corpus

[expressions including a word]

expressions including a word

[Intel i860 chip]

Intel i860 chip

[translates]

translates

[grammar coverage]

grammar coverage

[implications for discourse processing algorithms]

implications for discourse processing algorithms

[extending activity usually in a manner]

extending activity usually in a manner

[scope information of the correct structure]

scope information of the correct structure

[sophisticated units]

sophisticated units

[method acquiring data]

method acquiring data

[SMT algorithms]

SMT algorithms

[linguistic introspection]

linguistic introspection

[evaluation methodology]

evaluation methodology

[using conventional word-trigram recognition requiring transcription]

using conventional word-trigram recognition requiring transcription

[chart-based algorithms]

chart-based algorithms

[Category Cooccurrence Restrictions]

Category Cooccurrence Restrictions

[match accuracy]

match accuracy

[non-literal aspects such as communication procedures]

non-literal aspects such as communication procedures

[reformulation]

reformulation

[research on major new problems]

research on major new problems

[predicting concepts]

predicting concepts

[identification of 2-character names]

identification of 2-character names

[drastic generalization corresponding]

drastic generalization corresponding

[evaluating word sense WSD models]

evaluating word sense WSD models

[entropy word alignment algorithm]

entropy word alignment algorithm

[human performance]

human performance

[achieving performance to complex mixtures]

achieving performance to complex mixtures

[worst-case parsing time complexity]

worst-case parsing time complexity

[Complete distribution of knowledge]

Complete distribution of knowledge

[building a language model]

building a language model

[good predictors]

good predictors

[casting doubt]

casting doubt

[using expectations]

using expectations

[evidence from an ,000 features]

evidence from an ,000 features

[97% exact match accuracy]

97% exact match accuracy

[use in a common evaluation]

use in a common evaluation

[explicit use of tag contexts]

explicit use of tag contexts

[promising approach to Web-based writing]

promising approach to Web-based writing

[current discourse processing algorithms]

current discourse processing algorithms

[word classification]

word classification

[ONTOSCORE a system]

ONTOSCORE a system

[extending activity in a rightward manner]

extending activity in a rightward manner

[generic formalism for the combination]

generic formalism for the combination

[Lucy]

Lucy

[industry watch function]

industry watch function

[grammar formalisms such as systems]

grammar formalisms such as systems

[translation of language]

translation of language

[understanding of texts]

understanding of texts

[12,000 utterances of speech]

12,000 utterances of speech

[retrieving examples]

retrieving examples

[machine translation SMT]

machine translation SMT

[using associative processors APs]

using associative processors APs

[automatic evaluation measures]

automatic evaluation measures

[fundamentally different strategies utilizing knowledge-based mechanisms]

fundamentally different strategies utilizing knowledge-based mechanisms

[respectful words]

respectful words

[using additional features as evidence]

using additional features as evidence

[promising approach to academic writing]

promising approach to academic writing

[utterances of speech]

utterances of speech

[work harvesting English-Chinese bitexts]

work harvesting English-Chinese bitexts

[class-dependent interpolation]

class-dependent interpolation

[character recognition OCR model]

character recognition OCR model

[system non-monotonic reasoning]

system non-monotonic reasoning

[very simple sentence-plan-generator]

very simple sentence-plan-generator

[logical definition of grammars]

logical definition of grammars

[Network classifiers]

Network classifiers

[primarily on information from dictionaries MRD's]

primarily on information from dictionaries MRD's

[Information Bottleneck]

Information Bottleneck

[important task building the editor]

important task building the editor

[acquisition of information .g]

acquisition of information .g

[problems for readers]

problems for readers

[translations of the source text]

translations of the source text

[top-level prediction task]

top-level prediction task

[applying language]

applying language

[machine learning]

machine learning

[semantics of language applications]

semantics of language applications

[exploring bitexts in a larger volume]

exploring bitexts in a larger volume

[more commonly available resource]

more commonly available resource

[contributing matches]

contributing matches

[issue in Natural Language Processing]

issue in Natural Language Processing

[translation accuracy]

translation accuracy

[two kinds]

two kinds

[clues such as editing terms]

clues such as editing terms

[making analysis robust---an agenda-based scheduling parser]

making analysis robust---an agenda-based scheduling parser

[readers]

readers

[problems including vocabulary]

problems including vocabulary

[semantic equivalence classification]

semantic equivalence classification

[including jointly conditioning]

including jointly conditioning

[Hierarchical Phrasal Lexicon DHPL]

Hierarchical Phrasal Lexicon DHPL

[words from alignment models]

words from alignment models

[machine readable dictionaries]

machine readable dictionaries

[source interval]

source interval

[newspaper articles]

newspaper articles

[material the Canadian Hansards]

material the Canadian Hansards

[use in case-based MT]

use in case-based MT

[efficient parsing of Korean final language]

efficient parsing of Korean final language

[using tagging strategies]

using tagging strategies

[State]

State

[reach into other languages]

reach into other languages

[sentences in corpora]

sentences in corpora

[validity of such databases]

validity of such databases

[human reasoning]

human reasoning

[NE system]

NE system

[addressee identification face-to-face meetings]

addressee identification face-to-face meetings

[Prolog query optimisation]

Prolog query optimisation

[broad use of features]

broad use of features

[scope ambiguity]

scope ambiguity

[(BLEU]

(BLEU

[making exploratory corpus analysis]

making exploratory corpus analysis

[goal building that editor]

goal building that editor

[previous tagging result]

previous tagging result

[full analysis component]

full analysis component

[99% sentences]

99% sentences

[easy design]

easy design

[three types]

three types

[indifference]

indifference

[weak equivalence of the formalisms]

weak equivalence of the formalisms

[lexical rules]

lexical rules

[resulting expression]

resulting expression

[linguistic features]

linguistic features

[speech dialogue]

speech dialogue

[million word]

million word

[conveying the meaning]

conveying the meaning

[kinds of misunderstandings between speaker]

kinds of misunderstandings between speaker

[understanding system]

understanding system

[head-oriented notions]

head-oriented notions

[interesting combination of resources]

interesting combination of resources

[domain dependence]

domain dependence

[millions]

millions

[experiments corresponding]

experiments corresponding

[many forms]

many forms

[interactive disambiguation scheme]

interactive disambiguation scheme

[substring parsing]

substring parsing

[output selection]

output selection

[ways dealing with complexity]

ways dealing with complexity

[significant efficiency gains]

significant efficiency gains

[information .g]

information .g

[Finding the candidates]

Finding the candidates

[machine translation (SMT)]

machine translation (SMT)

[harvesting English-Chinese bitexts]

harvesting English-Chinese bitexts

[performance in applications]

performance in applications

[discovering sentences in corpora]

discovering sentences in corpora

[performance enhancing techniques including part]

performance enhancing techniques including part

[only edges with a semantic interpretation]

only edges with a semantic interpretation

[theory of reasoning]

theory of reasoning

[speech recognition of a language]

speech recognition of a language

[very small corpus words]

very small corpus words

[sense of much of the sentence]

sense of much of the sentence

[two standard outputs systems entity annotations]

two standard outputs systems entity annotations

[fundamentally different strategies]

fundamentally different strategies

[conglomeration of lexical rules]

conglomeration of lexical rules

[practical ways dealing]

practical ways dealing

[coherent speech recognition hypotheses]

coherent speech recognition hypotheses

[variations between the training corpus]

variations between the training corpus

[learning phrases longer than]

learning phrases longer than

[Grolier's]

Grolier's

[parallel corpus words]

parallel corpus words

[basics]

basics

[SI corpus]

SI corpus

[indices such as the time]

indices such as the time

[NLP components]

NLP components

[conditions for the use of expressions]

conditions for the use of expressions

[inferential theory 1988/]

inferential theory 1988/

[quantifier most]

quantifier most

[existing parser]

existing parser

[preliminary user evaluation]

preliminary user evaluation

[Korean Phonology Structure Grammar]

Korean Phonology Structure Grammar

[tiers]

tiers

[generating from redundancy]

generating from redundancy

[academic writing]

academic writing

[Lucy a portable understanding system]

Lucy a portable understanding system

[additional ,000 features]

additional ,000 features

[SENSEVAL-2 English sample task]

SENSEVAL-2 English sample task

[constructing a system speech recognition]

constructing a system speech recognition

[method visualizing]

method visualizing

[chart-based phrase structure]

chart-based phrase structure

[still valuable methods]

still valuable methods

[ii quality translation via word sense]

ii quality translation via word sense

[appealing alternative—in terms of simplicity]

appealing alternative—in terms of simplicity

[including part]

including part

[current capabilities]

current capabilities

[describing the processing in a discourse]

describing the processing in a discourse

[frequently good predictors of document descriptors]

frequently good predictors of document descriptors

[Isomorphic Grammars approach]

Isomorphic Grammars approach

[general-purpose language]

general-purpose language

[one superiority over methods]

one superiority over methods

[Complete lexical distribution]

Complete lexical distribution

[corresponding to arcs]

corresponding to arcs

[flexible inference method]

flexible inference method

[lexical distribution]

lexical distribution

[SENSEVAL-2 nouns]

SENSEVAL-2 nouns

[error correction of input]

error correction of input

[layout of an apartment]

layout of an apartment

[identification from a corpus]

identification from a corpus

[meaning of the sentence]

meaning of the sentence

[common evaluation]

common evaluation

[output of an probabilistic parser]

output of an probabilistic parser

[lexicon with entries]

lexicon with entries

[Techniques for training modules]

Techniques for training modules

[pooling]

pooling

[zero pronouns]

zero pronouns

[phonetic modelling]

phonetic modelling

[output sentence plans]

output sentence plans

[patterns in translation data]

patterns in translation data

[interactive method for speech understanding]

interactive method for speech understanding

[acquiring new stems from a word]

acquiring new stems from a word

[descriptors from phrases]

descriptors from phrases

[proposition p]

proposition p

[i Robust efficient parsing a verb]

i Robust efficient parsing a verb

[complexity of some]

complexity of some

[Chinese-to-English SMT model]

Chinese-to-English SMT model

[linguistic features of questions]

linguistic features of questions

[belief about the world]

belief about the world

[tree-to-tree transducer]

tree-to-tree transducer

[diverse pool]

diverse pool

[number of evaluations]

number of evaluations

[inherently a conglomeration]

inherently a conglomeration

[producing such phrases from a corpora]

producing such phrases from a corpora

[use of priors]

use of priors

[time solution]

time solution

[predictions of fragments]

predictions of fragments

[framework restrictive than earlier ones]

framework restrictive than earlier ones

[using interpolation of N-grams]

using interpolation of N-grams

[using a binary calculus notation]

using a binary calculus notation

[such MT]

such MT

[evidence from an additional features]

evidence from an additional features

[New results]

New results

[interpretation of language]

interpretation of language

[5 speed-up over a SUN]

5 speed-up over a SUN

[place of a semantics]

place of a semantics

[input expression]

input expression

[interactive question-answering system]

interactive question-answering system

[explicit development model]

explicit development model

[transfer component]

transfer component

[new part-of-speech tagger]

new part-of-speech tagger

[responses adaptive to individual users]

responses adaptive to individual users

[human understanding]

human understanding

[validity of such complex databases]

validity of such complex databases

[BU recognition system]

BU recognition system

[actual tape-recorded descriptions]

actual tape-recorded descriptions

[learning phrases words from alignment models]

learning phrases words from alignment models

[2 decision tree learning]

2 decision tree learning

[recovery technique]

recovery technique

[performance gains]

performance gains

[FERRET an question-answering system]

FERRET an question-answering system

[gathering statistics]

gathering statistics

[new inference method]

new inference method

[extraposition]

extraposition

[contextual clues]

contextual clues

[amount of internal lexical variation]

amount of internal lexical variation

[casting doubt on this assumption]

casting doubt on this assumption

[parsing language with case markers]

parsing language with case markers

[grammaticality 99% correct sentences]

grammaticality 99% correct sentences

[one superiority over existing methods]

one superiority over existing methods

[applying MT evaluation methods (BLEU]

applying MT evaluation methods (BLEU

[properties of tree-adjoining grammars]

properties of tree-adjoining grammars

[framework of collaborations]

framework of collaborations

[alternative—in terms]

alternative—in terms

[use in such MT CBMT]

use in such MT CBMT

[most dominant part]

most dominant part

[Common Coalition Language System]

Common Coalition Language System

[kinds of linguistic knowledge]

kinds of linguistic knowledge

[average precision over text]

average precision over text

[avoiding failures]

avoiding failures

[best single tagging result]

best single tagging result

[most probable morpheme sequence]

most probable morpheme sequence

[corpus of 100 words]

corpus of 100 words

[traditional word alignment techniques]

traditional word alignment techniques

[independently valuable NLP components]

independently valuable NLP components

[achieving performance to mixtures of techniques]

achieving performance to mixtures of techniques

[basis for the acquisition]

basis for the acquisition

[sets of rules]

sets of rules

[efficient parsing final language]

efficient parsing final language

[lexical gap]

lexical gap

[correct chain]

correct chain

[structures containing sets]

structures containing sets

[outputting unsegmented texts]

outputting unsegmented texts

[concept-based seeds]

concept-based seeds

[amount superior]

amount superior

[machine translation tests]

machine translation tests

[simple application]

simple application

[potential increase]

potential increase

[Conceptual Dependency Diagram]

Conceptual Dependency Diagram

[type of deterministic automaton]

type of deterministic automaton

[CCLINC Korean-to-English]

CCLINC Korean-to-English

[system with intelligent mobile agents]

system with intelligent mobile agents

[common ground]

common ground

[bearing]

bearing

[NE performance]

NE performance

[use of preceding tag contexts]

use of preceding tag contexts

[language texts e .g]

language texts e .g

[especially names]

especially names

[small corpus of words]

small corpus of words

[bilingual parallel corpora a available resource]

bilingual parallel corpora a available resource

[number of metaphor mappings]

number of metaphor mappings

[runtime]

runtime

[including unlimited vocabulary]

including unlimited vocabulary

[minor errors]

minor errors

[high-accuracy alignment models]

high-accuracy alignment models

[construction of lexica]

construction of lexica

[bilingual corpora a available resource]

bilingual corpora a available resource

[three methods .e LSA-based methods]

three methods .e LSA-based methods

[tagger use]

tagger use

[answers to everyday questions]

answers to everyday questions

[more data relevant to task]

more data relevant to task

[two phases]

two phases

[Hierarchical Acyclic Graph Kernel]

Hierarchical Acyclic Graph Kernel

[two new systems one]

two new systems one

[learning algorithm for identification]

learning algorithm for identification

[determining the meaning of input]

determining the meaning of input

[COMPLEX]

COMPLEX

[primarily on information from dictionaries]

primarily on information from dictionaries

[comparing the linguistic expressiveness]

comparing the linguistic expressiveness

[top-down scope information]

top-down scope information

[tagger explicit use]

tagger explicit use

[alphabets]

alphabets

[interaction requirements]

interaction requirements

[possible sentence plans]

possible sentence plans

[key factors including inherent limitations]

key factors including inherent limitations

[NIST MT evaluation software]

NIST MT evaluation software

[developing polynomial time approximations]

developing polynomial time approximations

[alignment quality]

alignment quality

[domain models]

domain models

[two problems a attributes]

two problems a attributes

[using a symbol]

using a symbol

[three components]

three components

[meetings using Network classifiers]

meetings using Network classifiers

[iii system development]

iii system development

[multiple meanings ambiguity]

multiple meanings ambiguity

[results of this experiment]

results of this experiment

[frame semantics paradigm]

frame semantics paradigm

[difficult SENSEVAL-2 nouns]

difficult SENSEVAL-2 nouns

[identifying words in newspapers]

identifying words in newspapers

[commercial systems]

commercial systems

[gathering information related]

gathering information related

[correcting homophone errors]

correcting homophone errors

[frequently individually good predictors]

frequently individually good predictors

[good baseline results]

good baseline results

[characteristics]

characteristics

[role of purpose]

role of purpose

[problems attributes]

problems attributes

[underspecified semantic representation]

underspecified semantic representation

[language data]

language data

[methods of natural language interfaces]

methods of natural language interfaces

[abstracting industry]

abstracting industry

[language essays]

language essays

[representational format for meaning]

representational format for meaning

[iv modeling of unknown word features]

iv modeling of unknown word features

[three capabilities]

three capabilities

[key features]

key features

[numerous aspects of communication]

numerous aspects of communication

[cluster number]

cluster number

[form language]

form language

[Using 40 utterances for adaptation]

Using 40 utterances for adaptation

[,000 features]

,000 features

[more precise description]

more precise description

[approach to machine translation]

approach to machine translation

[statistical MT systems]

statistical MT systems

[COMPLEX a computational lexicon]

COMPLEX a computational lexicon

[sequence in the pattern]

sequence in the pattern

[representations for document classification]

representations for document classification

[normal form]

normal form

[case study]

case study

[2 a response]

2 a response

[including jointly conditioning on words]

including jointly conditioning on words

[typical SMT models]

typical SMT models

[training data suitable]

training data suitable

[Grammars approach to MT]

Grammars approach to MT

[paradigm for speaker-independent SI training]

paradigm for speaker-independent SI training

[training set]

training set

[beginning]

beginning

[choice of structure]

choice of structure

[two contributions to large vocabulary]

two contributions to large vocabulary

[parsing toward meanings]

parsing toward meanings

[aggregation]

aggregation

[recent claims to the contrary]

recent claims to the contrary

[increasingly common claim]

increasingly common claim

[Language]

Language

[inherently a conglomeration of rules]

inherently a conglomeration of rules

[strong impact]

strong impact

[English analyst]

English analyst

[multiple corpora]

multiple corpora

[specification of an grammar model]

specification of an grammar model

[application of metrics]

application of metrics

[question-answering Q/A system]

question-answering Q/A system

[maximum a posteriori estimation]

maximum a posteriori estimation

[classifying undisambiguated SCF data]

classifying undisambiguated SCF data

[automatic Q/A applications]

automatic Q/A applications

[correlation of the measure]

correlation of the measure

[parser evaluation methods]

parser evaluation methods

[110 words]

110 words

[node with greater than one]

node with greater than one

[MT-based paraphrasing technique]

MT-based paraphrasing technique

[different information]

different information

[constructing a system]

constructing a system

[good predictors of document descriptors keywords]

good predictors of document descriptors keywords

['synchronous']

'synchronous'

[relevant aspects]

relevant aspects

[new disambiguation systems one]

new disambiguation systems one

[problems a attributes]

problems a attributes

[independent summarization approaches]

independent summarization approaches

[Graph]

Graph

[starting with a parallel corpus]

starting with a parallel corpus

[research on major problems]

research on major problems

[floppy]

floppy

[predicting top-level topic shifts]

predicting top-level topic shifts

[top-level boundaries]

top-level boundaries

[new methodology training SPoT]

new methodology training SPoT

[morpheme sequence for a input]

morpheme sequence for a input

[grammaticality 99% correct sentences equivalence]

grammaticality 99% correct sentences equivalence

[separate components the structure]

separate components the structure

[such edges as in treatments]

such edges as in treatments

[though rarely in the context]

though rarely in the context

[.0% improvement in the number]

.0% improvement in the number

[acquisition of information]

acquisition of information

[word-sense systems]

word-sense systems

[using a calculus notation]

using a calculus notation

[.1% a 45% reduction]

.1% a 45% reduction

[experiments corresponding to semantic domains]

experiments corresponding to semantic domains

[purposes a node]

purposes a node

[also a generalization]

also a generalization

[vagueness in annotation]

vagueness in annotation

[speech information]

speech information

[explicit information from machine-readable dictionaries]

explicit information from machine-readable dictionaries

[case of summarization]

case of summarization

[discourse through the operation]

discourse through the operation

[particular class]

particular class

[two stages]

two stages

[important bearing]

important bearing

[word n-grams]

word n-grams

[meaning representation]

meaning representation

[existing methods]

existing methods

[non-terminal symbol]

non-terminal symbol

[adaptive word formation]

adaptive word formation

[Recent advances in Speech Recognition technology]

Recent advances in Speech Recognition technology

[commercial systems outputting texts]

commercial systems outputting texts

[several factors]

several factors

[biological warfare]

biological warfare

[candidates for the homophone errors]

candidates for the homophone errors

[list of sentence plans]

list of sentence plans

[unsupervised technique learning morphology]

unsupervised technique learning morphology

[parallel corpora a more available resource]

parallel corpora a more available resource

[conglomeration of rules]

conglomeration of rules

[string with the performance]

string with the performance

[answering agents searching]

answering agents searching

[grammar tailoring]

grammar tailoring

[discovering parallel sentences in non-parallel corpora]

discovering parallel sentences in non-parallel corpora

[Parsing Model]

Parsing Model

[entity-oriented approach]

entity-oriented approach

[Common Coalition]

Common Coalition

[real-valued features a language model score]

real-valued features a language model score

[producing phrases from a word-aligned corpora]

producing phrases from a word-aligned corpora

[priors in loglinear models]

priors in loglinear models

[discovering sentences in non-parallel corpora]

discovering sentences in non-parallel corpora

[class-dependent interpolation of N-grams]

class-dependent interpolation of N-grams

[plausible interpretation from a chunk]

plausible interpretation from a chunk

[domain independent strategy]

domain independent strategy

[making corpus analysis accessible]

making corpus analysis accessible

[Coalition Language System at Lincoln Laboratory]

Coalition Language System at Lincoln Laboratory

[relationship between the grammatical formalisms]

relationship between the grammatical formalisms

[promising approach to computer-assisted writing]

promising approach to computer-assisted writing

[different aspects of the performance]

different aspects of the performance

[transcription of training data]

transcription of training data

[state proprietary stemmer]

state proprietary stemmer

[corpus exploration]

corpus exploration

[available resource]

available resource

[identifying patterns]

identifying patterns

[experience especially in the MUC-3 evaluation]

experience especially in the MUC-3 evaluation

[novice users]

novice users

[substantial improvement]

substantial improvement

[statistical techniques]

statistical techniques

[numerous non-literal aspects]

numerous non-literal aspects

[least 96% correct paraphrases]

least 96% correct paraphrases

[powerful computers]

powerful computers

[also a drastic generalization]

also a drastic generalization

[answering session]

answering session

[Research Methods]

Research Methods

[rule-based sentence planners]

rule-based sentence planners

[maximum likelihood method]

maximum likelihood method

[data collection paradigm]

data collection paradigm

[using corpora a available resource]

using corpora a available resource

[likewise automatic evaluation method]

likewise automatic evaluation method

[additional features of the tree]

additional features of the tree

[need for answers]

need for answers

[properties of grammars TAG]

properties of grammars TAG

[corpus consisting]

corpus consisting

[Syntactic analysis]

Syntactic analysis

[scoring system]

scoring system

[Graph HDAG]

Graph HDAG

[combining the styles]

combining the styles

[skill level]

skill level

[output of other tools]

output of other tools

[grammaticality sentences]

grammaticality sentences

[non-English stemmer]

non-English stemmer

[three methods constructing word vectors methods]

three methods constructing word vectors methods

[500 ,000 features]

500 ,000 features

[Sources of training data]

Sources of training data

[experiment in a series]

experiment in a series

[German corpus]

German corpus

[close correlation]

close correlation

[new word-sense systems one]

new word-sense systems one

[optical character recognition model]

optical character recognition model

[significantly less development time]

significantly less development time

[words from word-level alignment models]

words from word-level alignment models

[learning phrases longer than words]

learning phrases longer than words

[portion using a fake symbol]

portion using a fake symbol

[model of the global scene]

model of the global scene

[parse record]

parse record

[names without title]

names without title

[relatively simple means learning]

relatively simple means learning

[obviating the need]

obviating the need

[translingual reach into languages]

translingual reach into languages

[maximization]

maximization

[results on all applications]

results on all applications

[direct imitation of performance]

direct imitation of performance

[conversation analysis]

conversation analysis

[scoring alternative speech recognition hypotheses SRH]

scoring alternative speech recognition hypotheses SRH

[new type]

new type

[high performance]

high performance

[fully-connected first-order class grammar]

fully-connected first-order class grammar

[finding data relevant to task]

finding data relevant to task

[TAP-XL Analyst's Assistant]

TAP-XL Analyst's Assistant

[However the area]

However the area

[developing the grammars of the Source]

developing the grammars of the Source

[reflecting the relationship e .g]

reflecting the relationship e .g

[contributions to large vocabulary]

contributions to large vocabulary

[monitoring the collection for use]

monitoring the collection for use

[factors including limitations of MT architectures]

factors including limitations of MT architectures

[predicting the translation]

predicting the translation

[tracking the discourse]

tracking the discourse

[sophisticated annotation]

sophisticated annotation

[best hypothesis]

best hypothesis

[starting with a very small corpus]

starting with a very small corpus

[i Robust parsing a verb language]

i Robust parsing a verb language

[paradigm for SI training]

paradigm for SI training

[constructing a computational system speech recognition]

constructing a computational system speech recognition

[12 training speakers]

12 training speakers

[such as rewriting systems]

such as rewriting systems

[stemmer above]

stemmer above

[polarization]

polarization

[time approximations]

time approximations

[combination of various structures]

combination of various structures

[finite automaton]

finite automaton

[Chinese newspaper corpora]

Chinese newspaper corpora

[LSA-based methods]

LSA-based methods

[imitation of human performance]

imitation of human performance

[learning procedure]

learning procedure

[impediment]

impediment

[using Arabic information retrieval]

using Arabic information retrieval

[Following developments]

Following developments

[method identifying patterns]

method identifying patterns

[factor of 2]

factor of 2

[analysis robust---an scheduling parser]

analysis robust---an scheduling parser

[way improving system performance]

way improving system performance

[achieving performance]

achieving performance

[portion using a non-terminal symbol]

portion using a non-terminal symbol

[various training factors]

various training factors

[using large knowledgebases]

using large knowledgebases

[most expensive part]

most expensive part

[process translating a sentence]

process translating a sentence

[sentences parallel corpus]

sentences parallel corpus

[blocks pairs]

blocks pairs

[state of focus]

state of focus

[using parallel corpora]

using parallel corpora

[repair]

repair

[employing a generative phonological approach]

employing a generative phonological approach

[remote events]

remote events

[predictors of document descriptors keywords]

predictors of document descriptors keywords

[sentence utilizing examples]

sentence utilizing examples

[information e .g]

information e .g

[close correlation between the evaluation]

close correlation between the evaluation

[bootstrapping approach tagging]

bootstrapping approach tagging

[lexical weighting]

lexical weighting

[answering capability]

answering capability

[efficient parser for English]

efficient parser for English

[several field exercises]

several field exercises

[Model for language understanding]

Model for language understanding

[Dynamic Hierarchical Phrasal Lexicon]

Dynamic Hierarchical Phrasal Lexicon

[quality IE results]

quality IE results

[approach to statistical machine translation]

approach to statistical machine translation

[method for the reranking task]

method for the reranking task

[model of syntax-based machine translation]

model of syntax-based machine translation

[adding linguistic annotation]

adding linguistic annotation

[summarization quality of the system]

summarization quality of the system

[multiple candidates for understanding results]

multiple candidates for understanding results

[Traditional information]

Traditional information

[resolution]

resolution

[less development time]

less development time

[posing special problems]

posing special problems

[valuable NLP components]

valuable NLP components

[grammars of the Source]

grammars of the Source

[small 10K sentences corpus]

small 10K sentences corpus

[methods constructing word vectors .e]

methods constructing word vectors .e

[pool]

pool

[describing inflection]

describing inflection

[resolving pronouns]

resolving pronouns

[paraphrasing ambiguous sentences]

paraphrasing ambiguous sentences

[highest levels]

highest levels

[HPM]

HPM

[execution at the database]

execution at the database

[words with multiple meanings]

words with multiple meanings

[Paul]

Paul

[scarce resources]

scarce resources

[exploratory corpus analysis]

exploratory corpus analysis

[experience planning COMPLEX a lexicon]

experience planning COMPLEX a lexicon

[standard MT evaluation methods]

standard MT evaluation methods

[constructing a computational system]

constructing a computational system

[amount from the target speaker]

amount from the target speaker

[conglomeration of individual rules]

conglomeration of individual rules

[identifying unknown words in newspapers]

identifying unknown words in newspapers

[efficient parsing of Korean a verb]

efficient parsing of Korean a verb

[exploring bitexts]

exploring bitexts

[disambiguation tool]

disambiguation tool

[describing a case study]

describing a case study

[acquisition of constraints]

acquisition of constraints

[target word]

target word

[phrasal translation]

phrasal translation

[information redundancy]

information redundancy

[algorithm for the boosting approach]

algorithm for the boosting approach

[logical model in place]

logical model in place

[interpolation methods like log-linear interpolation]

interpolation methods like log-linear interpolation

[standard outputs from information extraction IE]

standard outputs from information extraction IE

[18 .6% improvement]

18 .6% improvement

[importantly Chinese-English texts]

importantly Chinese-English texts

[extracts of newswire text]

extracts of newswire text

[automatic extraction of translation lexicons]

automatic extraction of translation lexicons

[precise description of feature structures]

precise description of feature structures

[standard algorithms]

standard algorithms

[respect]

respect

[kinds of users]

kinds of users

[ongoing construction as basis]

ongoing construction as basis

[Acyclic Graph]

Acyclic Graph

[DARPA speech recognition]

DARPA speech recognition

[unification-based grammar parsing]

unification-based grammar parsing

[entering through asynchronous message passing]

entering through asynchronous message passing

[three words]

three words

[superset of L]

superset of L

[examples paraphrasing sentences]

examples paraphrasing sentences

[Stabler's formalization of Chomsky's program]

Stabler's formalization of Chomsky's program

[language generator]

language generator

[reach into languages]

reach into languages

[relational calculus notation]

relational calculus notation

[parsing mechanism]

parsing mechanism

[first test of an claim]

first test of an claim

[method discovering sentences]

method discovering sentences

[communication needs]

communication needs

[query optimisation in a database]

query optimisation in a database

[such as the time]

such as the time

[automatically acquiring stems from a word]

automatically acquiring stems from a word

[using machine learning techniques]

using machine learning techniques

[derivation tree]

derivation tree

[baseline of 54]

baseline of 54

[i Robust efficient parsing of Korean]

i Robust efficient parsing of Korean

[summarization approaches]

summarization approaches

[technology of powerful personal computers]

technology of powerful personal computers

[readers such as words]

readers such as words

[October test set]

October test set

[algorithms for unification of feature structures]

algorithms for unification of feature structures

[fuzzy quantifier most]

fuzzy quantifier most

[presence of a unknown]

presence of a unknown

[semantic constraints in domains]

semantic constraints in domains

[two new systems]

two new systems

[viewpoint]

viewpoint

[many grammar formalisms such as systems]

many grammar formalisms such as systems

[machine translation method]

machine translation method

[broad use including conditioning]

broad use including conditioning

[planning-based architecture with a variety]

planning-based architecture with a variety

[87 .5% agreement]

87 .5% agreement

[automatically training SPoT]

automatically training SPoT

[process in human understanding]

process in human understanding

[city]

city

[real-time language translation]

real-time language translation

[inferential theory]

inferential theory

[proposition with quantifiers]

proposition with quantifiers

[Several extensions]

Several extensions

[understanding system with agents]

understanding system with agents

[generalization of chart Parsing]

generalization of chart Parsing

[amount of variation slightly superior]

amount of variation slightly superior

[combining the styles in a system]

combining the styles in a system

[Rapid system development]

Rapid system development

[inflow of data]

inflow of data

[optical character recognition OCR model]

optical character recognition OCR model

[WSD accuracy]

WSD accuracy

[English-Spanish fragment]

English-Spanish fragment

[including limitations]

including limitations

[satisfying communication needs]

satisfying communication needs

[natural languages to semantics]

natural languages to semantics

[partial theories]

partial theories

[computational phonological system]

computational phonological system

[automatic extraction]

automatic extraction

[using word-trigram recognition requiring transcription]

using word-trigram recognition requiring transcription

[novel bootstrapping approach]

novel bootstrapping approach

[using bilingual parallel corpora]

using bilingual parallel corpora

[discourse processing algorithms]

discourse processing algorithms

[Implications automating aspects]

Implications automating aspects

[applying language understanding technology]

applying language understanding technology

[accuracy difference]

accuracy difference

[tree-adjoining grammars]

tree-adjoining grammars

[small corpus words]

small corpus words

[hypothesis corresponding]

hypothesis corresponding

[dependency insertion]

dependency insertion

[precise temporal adverbs]

precise temporal adverbs

[components the structure]

components the structure

[designing in particular]

designing in particular

[efficient parsing a verb language]

efficient parsing a verb language

[problems of SMT]

problems of SMT

[two outputs]

two outputs

[bigger a corpus]

bigger a corpus

[valuable general-purpose NLP components]

valuable general-purpose NLP components

[validity of entries]

validity of entries

[baseline sentence planners]

baseline sentence planners

[Czech-English machine translation system]

Czech-English machine translation system

[dynamic word formation]

dynamic word formation

[text mining]

text mining

[languages to semantics]

languages to semantics

[use of lexical features including conditioning]

use of lexical features including conditioning

[lexical-syntactic forms]

lexical-syntactic forms

[training reference]

training reference

[unification of feature structures]

unification of feature structures

[weak equivalence]

weak equivalence

[minutes]

minutes

[Graph HDAG Kernel]

Graph HDAG Kernel

[particular representational format]

particular representational format

[Complete distribution]

Complete distribution

[aiming for interpretations]

aiming for interpretations

[acquisition of constraints in domains]

acquisition of constraints in domains

[Chomsky's minimalist program]

Chomsky's minimalist program

[use of priors in models]

use of priors in models

[top-level subtopic boundaries]

top-level subtopic boundaries

[new systems one]

new systems one

[training speakers for SI recognition]

training speakers for SI recognition

[models of interaction]

models of interaction

[Korean knowledge base system]

Korean knowledge base system

[Dependency]

Dependency

[newspaper corpora]

newspaper corpora

[personal computers]

personal computers

[typical IBM MT system]

typical IBM MT system

[human-machine interactions]

human-machine interactions

[English translations]

English translations

[other areas]

other areas

[approach to writing]

approach to writing

[application beyond the confines]

application beyond the confines

[(given]

(given

[discovering sentences]

discovering sentences

[requiring transcription]

requiring transcription

[articulation of answers]

articulation of answers

[Following recent developments]

Following recent developments

[quality of summaries]

quality of summaries

[three components the structure]

three components the structure

[Arabic information retrieval]

Arabic information retrieval

[linguistic information]

linguistic information

[improvement over the baseline]

improvement over the baseline

[outputs from information extraction]

outputs from information extraction

[structure for elementary speech acts]

structure for elementary speech acts

[word sense disambiguation model]

word sense disambiguation model

[descriptions drawing a space]

descriptions drawing a space

[reranking task]

reranking task

[Markov chain model]

Markov chain model

[languages such as Czech]

languages such as Czech

[one author]

one author

[debate]

debate

[linking of entries to translation equivalents]

linking of entries to translation equivalents

[drastic generalization]

drastic generalization

[results on all four applications]

results on all four applications

[free word order]

free word order

[advantages of the strategies]

advantages of the strategies

[element in the design]

element in the design

[valency constraints]

valency constraints

[bootstrapping approach to Entity tagging]

bootstrapping approach to Entity tagging

[Using a word sense disambiguation model]

Using a word sense disambiguation model

[particular format]

particular format

[,449 word]

,449 word

[person's knowledge about knowledge]

person's knowledge about knowledge

[pairs of probabilistic context-free grammars]

pairs of probabilistic context-free grammars

[distance]

distance

[systems one]

systems one

[treatment of syntax]

treatment of syntax

[underlying computation model]

underlying computation model

[sounding dialog systems]

sounding dialog systems

[relationship between the two formalisms]

relationship between the two formalisms

[problems attributes such as size perspective-taking]

problems attributes such as size perspective-taking

[28 ,449 word tokens]

28 ,449 word tokens

[dictionary on the basis]

dictionary on the basis

[styles performing inference in networks]

styles performing inference in networks

[such edges]

such edges

[speaker adaptation SA]

speaker adaptation SA

[using these models]

using these models

[guidance]

guidance

[instance to the task]

instance to the task

[phrase-based translation models]

phrase-based translation models

[improvement in precision over text]

improvement in precision over text

[NE rules]

NE rules

[Unification Categorial Grammar]

Unification Categorial Grammar

[language texts]

language texts

[roughly corresponding to arcs]

roughly corresponding to arcs

[error reduction]

error reduction

[case-based MT]

case-based MT

[method adding linguistic annotation]

method adding linguistic annotation

[simple sentence-plan-generator]

simple sentence-plan-generator

[sounding dialog systems within reach]

sounding dialog systems within reach

[current syntax formalisms]

current syntax formalisms

[use in MT CBMT]

use in MT CBMT

[linking of lexical entries]

linking of lexical entries

[machine translation evaluation measures]

machine translation evaluation measures

[Acyclic Graph HDAG Kernel]

Acyclic Graph HDAG Kernel

[analyzing sentences]

analyzing sentences

[Unification Grammar]

Unification Grammar

[good choice]

good choice

[vehicle]

vehicle

[explanation of discourse phenomena]

explanation of discourse phenomena

[authoritative text collection]

authoritative text collection

[scoring speech recognition hypotheses]

scoring speech recognition hypotheses

[words like sentence]

words like sentence

[global meaning]

global meaning

[bottom-up pattern-matching parser]

bottom-up pattern-matching parser

[flexible architecture]

flexible architecture

[Dynamic Phrasal Lexicon DHPL]

Dynamic Phrasal Lexicon DHPL

[algorithm acquiring stems]

algorithm acquiring stems

[parse of a sentence]

parse of a sentence

[.0% improvement over baseline system]

.0% improvement over baseline system

[Parsing Model HPM]

Parsing Model HPM

[extensions of this idea]

extensions of this idea

[different aspects including the influence]

different aspects including the influence

[only edges with a valid interpretation]

only edges with a valid interpretation

[multimedia data]

multimedia data

[corpus 100 words]

corpus 100 words

[voice system]

voice system

[approach to computer-assisted writing]

approach to computer-assisted writing

[frequently predictors of document descriptors keywords]

frequently predictors of document descriptors keywords

[challenges integrating Q/A applications]

challenges integrating Q/A applications

[model of natural language]

model of natural language

[Hierarchical Phrasal Lexicon [Zernik88]]

Hierarchical Phrasal Lexicon [Zernik88]

[English understanding system]

English understanding system

[word formation]

word formation

[very small parallel corpus]

very small parallel corpus

[Following developments in the evaluation]

Following developments in the evaluation

[coverage English grammar]

coverage English grammar

[issue in Language Processing]

issue in Language Processing

[using additional features]

using additional features

[terms on feature selection methods]

terms on feature selection methods

[engine machine translation system]

engine machine translation system

[Model for natural language understanding]

Model for natural language understanding

[one element in the design]

one element in the design

[standard outputs]

standard outputs

[identifying systematic patterns]

identifying systematic patterns

[Chinese newspapers]

Chinese newspapers

[2-character Chinese names]

2-character Chinese names

[combiner with decisions]

combiner with decisions

[top-level topic shifts]

top-level topic shifts

[use including jointly conditioning]

use including jointly conditioning

[first experiment]

first experiment

[imitation]

imitation

[linguistic representation the Phrasal Lexicon]

linguistic representation the Phrasal Lexicon

[protocols from answers to questions]

protocols from answers to questions

[collecting statistics]

collecting statistics

[cooperative responses adaptive to users]

cooperative responses adaptive to users

[protocols from people's answers]

protocols from people's answers

[terms of Grammars]

terms of Grammars

[cohesion-based approach alone]

cohesion-based approach alone

[combination of resources]

combination of resources

[different strategies utilizing knowledge-based mechanisms]

different strategies utilizing knowledge-based mechanisms

[syntactic disambiguation]

syntactic disambiguation

[several factors behind this surprising finding]

several factors behind this surprising finding

[event-based extractive summarization approaches]

event-based extractive summarization approaches

[case markers]

case markers

[paraphrasing sentences]

paraphrasing sentences

[evaluation disambiguation algorithms]

evaluation disambiguation algorithms

[University of California]

University of California

[automating aspects]

automating aspects

[space including a variety]

space including a variety

[shortcomings of strategies]

shortcomings of strategies

[proposition with fuzzy quantifiers]

proposition with fuzzy quantifiers

[,610 sentences]

,610 sentences

[hand-built resources]

hand-built resources

[independent event-based mutli-document summarization approaches]

independent event-based mutli-document summarization approaches

[inferential theory for language presuppositions]

inferential theory for language presuppositions

[view for reference]

view for reference

[few noun]

few noun

[now old Resource Management corpus]

now old Resource Management corpus

[TDMT the dominant part]

TDMT the dominant part

[information redundancy in input]

information redundancy in input

[corpora a resource]

corpora a resource

[tree-adjoining grammars TAG]

tree-adjoining grammars TAG

[parallel corpora a available resource]

parallel corpora a available resource

[primarily on information from machine-readable dictionaries]

primarily on information from machine-readable dictionaries

[MAP estimation]

MAP estimation

[coordinate conjunction]

coordinate conjunction

[reproducability]

reproducability

[40 utterances]

40 utterances

[least 99% sentences]

least 99% sentences

[Resource Management corpus]

Resource Management corpus

[top-down scope information of the structure]

top-down scope information of the structure

[NE e .g]

NE e .g

[intelligent systems]

intelligent systems

[validity]

validity

[new technology]

new technology

[preceding]

preceding

[using these tree-based models]

using these tree-based models

[conventional SMT models]

conventional SMT models

[effective way]

effective way

[Transfer System SimTran]

Transfer System SimTran

[improvements in performance]

improvements in performance

[ideal aiming]

ideal aiming

[more than 99% accuracy]

more than 99% accuracy

[function words]

function words

[knowledge level]

knowledge level

[effect of polysemy]

effect of polysemy

[aspects such as robust communication procedures]

aspects such as robust communication procedures

[scheduling parser a recovery technique]

scheduling parser a recovery technique

[primarily passing some issues]

primarily passing some issues

[assigning ranks]

assigning ranks

[including a variety]

including a variety

[grammaticality sentences equivalence in meaning]

grammaticality sentences equivalence in meaning

[various structures]

various structures

[relative effects]

relative effects

[PERSON NE]

PERSON NE

[separation margin]

separation margin

[automatically identifying argument structures]

automatically identifying argument structures

[novel method adding annotation]

novel method adding annotation

[SI training]

SI training

[results on addressee identification]

results on addressee identification

[human translations]

human translations

[light of various functions]

light of various functions

[generative character recognition model]

generative character recognition model

[translation output]

translation output

[robust method]

robust method

[distinct phases]

distinct phases

[human communication]

human communication

[research topics]

research topics

[POS tagger]

POS tagger

[only edges with a interpretation]

only edges with a interpretation

[MT CBMT]

MT CBMT

[three methods .e]

three methods .e

[Manual acquisition]

Manual acquisition

[language with case markers]

language with case markers

[pairs of grammars]

pairs of grammars

[subjective human judgments]

subjective human judgments

[clear evidence]

clear evidence

[such as size perspective-taking in reference]

such as size perspective-taking in reference

[Bayesian Network classifiers]

Bayesian Network classifiers

[node with greater]

node with greater

[new tagger use]

new tagger use

[.24% accuracy an error reduction]

.24% accuracy an error reduction

[such as robust communication procedures]

such as robust communication procedures

[spatial constraints]

spatial constraints

[domain independent acoustic models]

domain independent acoustic models

[human communication needs]

human communication needs

[combining the two styles]

combining the two styles

[grammaticality 99% sentences]

grammaticality 99% sentences

[subcategorization frame SCF]

subcategorization frame SCF

[Robust efficient parsing a verb language]

Robust efficient parsing a verb language

[retrieval of phrases]

retrieval of phrases

[methods constructing word vectors methods]

methods constructing word vectors methods

[high-accuracy word-level alignment models]

high-accuracy word-level alignment models

[Hierarchical Acyclic Graph]

Hierarchical Acyclic Graph

[generalization corresponding in a chart]

generalization corresponding in a chart

[segment methods in terms]

segment methods in terms

[likelihood method]

likelihood method

[qualitative constraints from the text]

qualitative constraints from the text

[using corpora a commonly available resource]

using corpora a commonly available resource

[covering that gap]

covering that gap

[k-means algorithm]

k-means algorithm

[semantic frame encoding the meaning]

semantic frame encoding the meaning

[predicting linguistic concepts]

predicting linguistic concepts

[encoding the meaning]

encoding the meaning

[case study avoiding failures]

case study avoiding failures

[new corpus exploration tool]

new corpus exploration tool

[Robust parsing of Korean language]

Robust parsing of Korean language

[manual word alignments]

manual word alignments

[certain classes of algorithms]

certain classes of algorithms

[approximation of the analysis]

approximation of the analysis

[evaluating a Chinese-to-English SMT model]

evaluating a Chinese-to-English SMT model

[especially in the MUC-3 evaluation]

especially in the MUC-3 evaluation

[parser for English]

parser for English

[Manual acquisition of constraints]

Manual acquisition of constraints

[notions of valency]

notions of valency

[distribution of knowledge]

distribution of knowledge

[producing phrases]

producing phrases

[new way]

new way

[encoding the full meaning]

encoding the full meaning

[novel Model]

novel Model

[translates in another language]

translates in another language

[multiple parsing strategies]

multiple parsing strategies

[collection of utterances from sites]

collection of utterances from sites

[pattern-matching]

pattern-matching

[output sentence]

output sentence

[corpus consisting of 126 sentences]

corpus consisting of 126 sentences

[using an bilingual corpus]

using an bilingual corpus

[identification of paraphrases]

identification of paraphrases

[framework of Grammars]

framework of Grammars

[i parsing a verb]

i parsing a verb

[subcategorization]

subcategorization

[building that editor]

building that editor

[WSD accuracy]

WSD accuracy

[missing words]

missing words

[system performance]

system performance

[techniques making analysis]

techniques making analysis

[advantages from the point]

advantages from the point

[35 .0% improvement]

35 .0% improvement

[building upon the notions of valency]

building upon the notions of valency

[only the topmost]

only the topmost

[96% correct paraphrases]

96% correct paraphrases

[Using only 40 utterances]

Using only 40 utterances

[pairs of context-free grammars]

pairs of context-free grammars

[Navy messages]

Navy messages

[human-ranked sentence plan]

human-ranked sentence plan

[language system domains]

language system domains

[language generation community]

language generation community

[failures of reference]

failures of reference

[hardness results for the class NP]

hardness results for the class NP

[word-senses of words]

word-senses of words

[Multi-site]

Multi-site

[following speech recognition applications]

following speech recognition applications

[reference about the discourse]

reference about the discourse

[dialogue duration for users]

dialogue duration for users

[dimensions skill level]

dimensions skill level

[examples similar]

examples similar

[including a frequent word]

including a frequent word

[key factors including limitations]

key factors including limitations

[three techniques making analysis]

three techniques making analysis

[grammars working]

grammars working

[scene]

scene

[programming language]

programming language

[head-orientation]

head-orientation

[use of parsing strategies]

use of parsing strategies

[detecting speech repairs]

detecting speech repairs

[grounding acts]

grounding acts

[97 .24% accuracy]

97 .24% accuracy

[training factors]

training factors

[much simpler set]

much simpler set

[standard models]

standard models

[NLP techniques]

NLP techniques

[13% decrease]

13% decrease

[many classical formalisms]

many classical formalisms

[lexical cross-relations]

lexical cross-relations

[access]

access

[three words from high-accuracy alignment models]

three words from high-accuracy alignment models

[bootstrapping approach]

bootstrapping approach

[possible span]

possible span

[information from machine-readable dictionaries MRD's]

information from machine-readable dictionaries MRD's

[dependency pointers]

dependency pointers

[substitutions for these elements]

substitutions for these elements

[ensemble methods]

ensemble methods

[effect using an out-of-domain corpus]

effect using an out-of-domain corpus

[reflecting the social relationship social ranks]

reflecting the social relationship social ranks

[stochastic output selection]

stochastic output selection

[including conditioning]

including conditioning

[appropriate pints]

appropriate pints

[evaluation measures for machine translation]

evaluation measures for machine translation

[length-based criterion]

length-based criterion

[domains tutorial dialogue]

domains tutorial dialogue

[two disambiguation systems]

two disambiguation systems

[COMPLEX a lexicon]

COMPLEX a lexicon

[instantiation of the sense]

instantiation of the sense

[Machine Translation SMT]

Machine Translation SMT

[harvesting bitexts from the Web]

harvesting bitexts from the Web

[exploring English-Chinese bitexts]

exploring English-Chinese bitexts

[89 .75% F-measure a decrease]

89 .75% F-measure a decrease

[engine machine]

engine machine

[strength for each]

strength for each

[for scoring sets]

for scoring sets

[generative model of language]

generative model of language

[i efficient parsing final language]

i efficient parsing final language

[valid interpretation]

valid interpretation

[rule-reduction]

rule-reduction

[degree]

degree

[3 indicators for the prediction task]

3 indicators for the prediction task

[progressing from generation into the output]

progressing from generation into the output

[Most state-of-the-art evaluation measures]

Most state-of-the-art evaluation measures

[operational semantics of language applications]

operational semantics of language applications

[question-answering]

question-answering

[experience planning COMPLEX]

experience planning COMPLEX

[problem combining several language models]

problem combining several language models

[formalism for the combination]

formalism for the combination

[FlexP a pattern-matching parser]

FlexP a pattern-matching parser

[instead of pairs]

instead of pairs

[Robust efficient parsing final language]

Robust efficient parsing final language

[Robust efficient parsing of Korean language]

Robust efficient parsing of Korean language

[attributes of the entities]

attributes of the entities

[learning morphology]

learning morphology

[virtue]

virtue

[miscommunication reference]

miscommunication reference

[amount slightly superior]

amount slightly superior

[Horn logic]

Horn logic

[such discourse phenomena]

such discourse phenomena

[distinct datasets]

distinct datasets

[hypothesis corresponding to the LM]

hypothesis corresponding to the LM

[iii system development via acquisition]

iii system development via acquisition

[class of words]

class of words

[building upon the notions]

building upon the notions

[amount of lexical variation]

amount of lexical variation

[empirical results casting doubt]

empirical results casting doubt

[overheads]

overheads

[i parsing language]

i parsing language

[segment boundaries in multiparty dialogue]

segment boundaries in multiparty dialogue

[parsing language with overt case markers]

parsing language with overt case markers

[evaluation results involving extraction]

evaluation results involving extraction

[cross-relations]

cross-relations

[including limitations of current MT architectures]

including limitations of current MT architectures

[novel data structure]

novel data structure

[Tree Adjoining Grammars]

Tree Adjoining Grammars

[test of an common claim]

test of an common claim

[instantiation of clauses in a program]

instantiation of clauses in a program

[analysis for a class]

analysis for a class

[important issue in Language Processing]

important issue in Language Processing

[language families]

language families

[producing such phrases]

producing such phrases

[approach to computer-assisted academic writing]

approach to computer-assisted academic writing

[exploitation of limited domain semantics]

exploitation of limited domain semantics

[n-gram models]

n-gram models

[polysemic verbs]

polysemic verbs

[work harvesting bitexts from the Web]

work harvesting bitexts from the Web

[sites]

sites

[Implications automating certain aspects]

Implications automating certain aspects

[using a statistical POS tagger]

using a statistical POS tagger

[workability of the mechanism]

workability of the mechanism

[Traditional information retrieval]

Traditional information retrieval

[old Resource Management corpus]

old Resource Management corpus

[methods methods]

methods methods

[levels of chunks]

levels of chunks

[outputting texts]

outputting texts

[novel bootstrapping approach to Entity]

novel bootstrapping approach to Entity

[generalization of chart Parsing instantiation]

generalization of chart Parsing instantiation

[problems for readers such as words]

problems for readers such as words

[traditional generative approach]

traditional generative approach

[partial instantiation of clauses]

partial instantiation of clauses

[using parallel corpora a resource]

using parallel corpora a resource

[heuristically-produced sentences]

heuristically-produced sentences

[prediction task]

prediction task

[Human-Machine Communication]

Human-Machine Communication

[generative probabilistic model]

generative probabilistic model

[describing the processing]

describing the processing

[word-semantic information e]

word-semantic information e

[integrating Q/A applications into real-world environments]

integrating Q/A applications into real-world environments

[collecting statistics on co-occurrence patterns]

collecting statistics on co-occurrence patterns

[study of expressions]

study of expressions

[PERSON]

PERSON

[new theory]

new theory

[great effectiveness]

great effectiveness

[word-trigram recognition requiring transcription]

word-trigram recognition requiring transcription

[spatial attributes]

spatial attributes

[applying evaluation techniques to the output]

applying evaluation techniques to the output

[global behavior]

global behavior

[robust recognition of input]

robust recognition of input

[distance constraints]

distance constraints

[method discovering sentences in corpora]

method discovering sentences in corpora

[identifying argument structures]

identifying argument structures

[correcting errors]

correcting errors

[model analyzing sentences]

model analyzing sentences

[predictors]

predictors

[part of a study of expressions]

part of a study of expressions

[state-of-the-art word sense]

state-of-the-art word sense

[styles performing inference]

styles performing inference

[parsing capability]

parsing capability

[learning phrases three words]

learning phrases three words

[model for stochastic output selection]

model for stochastic output selection

[more data relevant]

more data relevant

[orientation]

orientation

[standard MT evaluation methods (BLEU]

standard MT evaluation methods (BLEU

[/Soames]

/Soames

[light of rhetorical functions]

light of rhetorical functions

[transcription errors]

transcription errors

[disjunctive form]

disjunctive form

[interaction with knowledge about dialogs]

interaction with knowledge about dialogs

[class of miscommunication reference problems]

class of miscommunication reference problems

[speed-up element in the design]

speed-up element in the design

[acquiring stems from a word]

acquiring stems from a word

[corpus of .284 SRHs]

corpus of .284 SRHs

[relatively simple means]

relatively simple means

[approaches to WSI]

approaches to WSI

[model of machine translation MT]

model of machine translation MT

[Grammars for Machine Translation]

Grammars for Machine Translation

[system development]

system development

[computational algorithms for unification]

computational algorithms for unification

[building a non-English stemmer]

building a non-English stemmer

[computational issues]

computational issues

[passing some computational issues]

passing some computational issues

[ongoing construction]

ongoing construction

[Robust efficient parsing]

Robust efficient parsing

[little gain from information]

little gain from information

[size]

size

[i Robust efficient parsing final language]

i Robust efficient parsing final language

[several factors behind this finding]

several factors behind this finding

[processing examples]

processing examples

[wide coverage]

wide coverage

[Hierarchical Graph HDAG]

Hierarchical Graph HDAG

[methods of traditional language interfaces]

methods of traditional language interfaces

[distribution of behaviors]

distribution of behaviors

[generic formalism]

generic formalism

[descriptions in Japanese]

descriptions in Japanese

[combinatorics]

combinatorics

[real applications]

real applications

[New results on all four applications]

New results on all four applications

[corresponding to domains]

corresponding to domains

[harvesting bitexts]

harvesting bitexts

[employing n-gram models for Thai prediction]

employing n-gram models for Thai prediction

[definition of grammars]

definition of grammars

[use of various features]

use of various features

[problems attributes such as size]

problems attributes such as size

[.1% a reduction]

.1% a reduction

[small number]

small number

[explicit information]

explicit information

[transition graphs for a special type]

transition graphs for a special type

[natural language applications]

natural language applications

[analysis of answers to questions]

analysis of answers to questions

[making]

making

[scoring alternative speech recognition hypotheses]

scoring alternative speech recognition hypotheses

[question-answering system]

question-answering system

[skill level to the system]

skill level to the system

[informational goals]

informational goals

[stochastic transducer]

stochastic transducer

[computational theory of human reasoning]

computational theory of human reasoning

[word-trigram recognition]

word-trigram recognition

[lexical approach alone]

lexical approach alone

[such as the dominance distribution]

such as the dominance distribution

[corner parsing]

corner parsing

[common speculative claim]

common speculative claim

[using corpora]

using corpora

[strategies one utilizing knowledge-based mechanisms]

strategies one utilizing knowledge-based mechanisms

[exact match accuracy]

exact match accuracy

[topic shifts]

topic shifts

[social relationship social ranks]

social relationship social ranks

[lexical distribution of grammatical knowledge]

lexical distribution of grammatical knowledge

[subcategorization frame]

subcategorization frame

[i Robust parsing a verb]

i Robust parsing a verb

[clustering subcategorization frame distributions]

clustering subcategorization frame distributions

[simultaneously using memory]

simultaneously using memory

[broad-coverage parser]

broad-coverage parser

[inevitable inferences]

inevitable inferences

[ongoing construction as reliable basis]

ongoing construction as reliable basis

[free text]

free text

[language in natural settings]

language in natural settings

[using the SI corpus]

using the SI corpus

[quasi-destructive scheme's ability]

quasi-destructive scheme's ability

[new word-sense systems]

new word-sense systems

[small-sized corpus]

small-sized corpus

[grammar formalism Phonology Structure Grammar]

grammar formalism Phonology Structure Grammar

[formal analysis for a class]

formal analysis for a class

[other areas of language processing]

other areas of language processing

[more frequently good predictors]

more frequently good predictors

[discourse constraint]

discourse constraint

[processing time]

processing time

[personnel]

personnel

[sequence of morphemes]

sequence of morphemes

[particular format for meaning]

particular format for meaning

[improvement of 22-38% in precision]

improvement of 22-38% in precision

[coverage than accounts]

coverage than accounts

[large inflow of data]

large inflow of data

[distribution of forms]

distribution of forms

[bilingual corpora a more available resource]

bilingual corpora a more available resource

[treelet]

treelet

[certain natural sublanguages]

certain natural sublanguages

[/Gazdar]

/Gazdar

[graphics displays]

graphics displays

[field exercises]

field exercises

[statistical class grammar]

statistical class grammar

[only the topmost of the edges]

only the topmost of the edges

[two decades]

two decades

[Markov models]

Markov models

[lexical distribution of knowledge]

lexical distribution of knowledge

[complex mixtures]

complex mixtures

[information from dictionaries MRD's]

information from dictionaries MRD's

[estimation of continuous Markov models]

estimation of continuous Markov models

[mathematical formalism for the combination]

mathematical formalism for the combination

[words with meanings ambiguity]

words with meanings ambiguity

[little gain]

little gain

[syntactic analysis robust---an scheduling parser]

syntactic analysis robust---an scheduling parser

[efficient parsing of Korean language]

efficient parsing of Korean language

[broad use]

broad use

[use of features including jointly conditioning]

use of features including jointly conditioning

[inherently a conglomeration of lexical rules]

inherently a conglomeration of lexical rules

[subject of recent debate]

subject of recent debate

[U .S]

U .S

[fundamentally different strategies one utilizing mechanisms]

fundamentally different strategies one utilizing mechanisms

[non-deterministic parsing choices]

non-deterministic parsing choices

[Several advantages]

Several advantages

[analysis of the coordinate sentences]

analysis of the coordinate sentences

[knowledge concerning translation equivalence]

knowledge concerning translation equivalence

[linguistic databases]

linguistic databases

[tape-recorded descriptions]

tape-recorded descriptions

[using a model]

using a model

[showing the WSD accuracy lower]

showing the WSD accuracy lower

[decisions using the reference]

decisions using the reference

[picking the hypothesis]

picking the hypothesis

[translation task]

translation task

[correcting Japanese homophone errors]

correcting Japanese homophone errors

[experiments with substitution costs]

experiments with substitution costs

[decrease in F-measure error]

decrease in F-measure error

[.5% word error rate]

.5% word error rate

[acquiring stems from a word corpus]

acquiring stems from a word corpus

[depending on the type]

depending on the type

[two word-sense systems one]

two word-sense systems one

[binary calculus notation]

binary calculus notation

[best existing parsing models]

best existing parsing models

[possibility]

possibility

[eliciting subjective judgments]

eliciting subjective judgments

[context of dialog systems]

context of dialog systems

[robust probabilistic parsing models]

robust probabilistic parsing models

[chart Parsing partial instantiation]

chart Parsing partial instantiation

[statistical MT architectures]

statistical MT architectures

[term weighting assignment paradigm]

term weighting assignment paradigm

[methods of language interfaces]

methods of language interfaces

[concerning translation equivalence of expressions]

concerning translation equivalence of expressions

[event detection]

event detection

[acquiring data]

acquiring data

[learning of phrase translations from alignments]

learning of phrase translations from alignments

[probabilistic grammars working]

probabilistic grammars working

[parallel corpora a commonly available resource]

parallel corpora a commonly available resource

[voting-]

voting-

[simple explanation for the nature]

simple explanation for the nature

[certain sublanguages]

certain sublanguages

[identifying subtopic boundaries]

identifying subtopic boundaries

[word sense disambiguation performance]

word sense disambiguation performance

[construction as reliable basis]

construction as reliable basis

[methodology training SPoT]

methodology training SPoT

[symmetric patterns]

symmetric patterns

[passing some issues]

passing some issues

[problem of word sense]

problem of word sense

[dialogue act]

dialogue act

[MLE reestimation]

MLE reestimation

[validity of databases]

validity of databases

[explanation for the presuppositional nature]

explanation for the presuppositional nature

[use in such case-based MT]

use in such case-based MT

[strongly biasing parsing]

strongly biasing parsing

[surface representation]

surface representation

[amount of variation]

amount of variation

[Dynamic Phrasal Lexicon [Zernik88]]

Dynamic Phrasal Lexicon [Zernik88]

[10 .4M words]

10 .4M words

[SmTA check]

SmTA check

[advantages like reproducability]

advantages like reproducability

[factors including inherent limitations]

factors including inherent limitations

[exploiting a large corpus]

exploiting a large corpus

[translation use]

translation use

[training modules]

training modules

[complex linguistic databases]

complex linguistic databases

[test of an claim]

test of an claim

[elementary speech acts]

elementary speech acts

[phrase structure parsing]

phrase structure parsing

[independent mutli-document summarization approaches]

independent mutli-document summarization approaches

[theory of human plausible reasoning]

theory of human plausible reasoning

[ACE corpora]

ACE corpora

[necessary conditions]

necessary conditions

[addition the language model]

addition the language model

[dependency insertion grammar]

dependency insertion grammar, dependency insertion grammars

[open-domain question]

open-domain question

[question-answering Q/A]

question-answering Q/A

[fashion]

fashion

[corpus of SRHs]

corpus of SRHs

[two learners]

two learners

[sentences corpus as training resources]

sentences corpus as training resources

[strong discourse effect]

strong discourse effect

[co-occurrence patterns]

co-occurrence patterns

[learning methodology]

learning methodology

[approach alone]

approach alone

[complicated knowledgebases]

complicated knowledgebases

[separation]

separation

[automatically gathering a large number]

automatically gathering a large number

[acquisition the construction]

acquisition the construction

[simple technique learning morphology]

simple technique learning morphology

[parsing accuracy]

parsing accuracy

[minor extensions]

minor extensions

[top sentence plan]

top sentence plan

[empirical test of an claim]

empirical test of an claim

[word segmentation system]

word segmentation system

[part-of-speech tagger use]

part-of-speech tagger use

[large newspaper corpora]

large newspaper corpora

[motivation]

motivation

[independent extractive mutli-document summarization approaches]

independent extractive mutli-document summarization approaches

[primarily on explicit information]

primarily on explicit information

[building translation systems]

building translation systems

[applying general-purpose language]

applying general-purpose language

[computer text generation]

computer text generation

[rhetorical functions]

rhetorical functions

[several language models]

several language models

[F-measure a 13% relative decrease]

F-measure a 13% relative decrease

[text through transformation]

text through transformation

[evaluating a representative SMT model]

evaluating a representative SMT model

[fewer equivalent readings]

fewer equivalent readings

[corpus of English translations]

corpus of English translations

[though in the context]

though in the context

[resource logic]

resource logic

[pairs]

pairs

[first test]

first test

[sets( DoPS]

sets( DoPS

[distinct tasks]

distinct tasks

[analysis of move structures]

analysis of move structures

[thereby reducing the amount]

thereby reducing the amount

[bunsetsu]

bunsetsu

[also a generalization roughly corresponding]

also a generalization roughly corresponding

[right translation]

right translation

[properties of noun phrases]

properties of noun phrases

[Network]

Network

[factors behind this surprising finding]

factors behind this surprising finding

[10K sentences corpus]

10K sentences corpus

[translating a sentence]

translating a sentence

[automatically training modules]

automatically training modules

[controlling the order]

controlling the order

[multimedia articulation]

multimedia articulation

[corpora a commonly available resource]

corpora a commonly available resource

[time solution for any]

time solution for any

[consisting of ,610 sentences]

consisting of ,610 sentences

[integrating automatic Q/A applications]

integrating automatic Q/A applications

[adequate explanation]

adequate explanation

[Phrasal Lexicon DHPL [Zernik88]]

Phrasal Lexicon DHPL [Zernik88]

[conversational vehicle]

conversational vehicle

[Bayesian Network]

Bayesian Network

[automatically acquiring stems]

automatically acquiring stems

[using information retrieval]

using information retrieval

[Role System]

Role System

[using a POS tagger in conjunction]

using a POS tagger in conjunction

[experimental results on all four applications]

experimental results on all four applications

[three minutes]

three minutes

[two domains tutorial dialogue]

two domains tutorial dialogue

[speed-up element]

speed-up element

[understanding system with mobile agents]

understanding system with mobile agents

[including limitations of statistical MT architectures]

including limitations of statistical MT architectures

[pints in MT]

pints in MT

[cues such as cue phrases]

cues such as cue phrases

[LHIP]

LHIP

[new paradigm]

new paradigm

[alternative—in terms of simplicity]

alternative—in terms of simplicity

[message-passing infrastructure]

message-passing infrastructure

[constructing a computational phonological system]

constructing a computational phonological system

[linguistic parser/generator a transfer component]

linguistic parser/generator a transfer component

[using a logical model]

using a logical model

[theory for language presuppositions]

theory for language presuppositions

[adding annotation to corpora]

adding annotation to corpora

[model of the scene]

model of the scene

[baseline results]

baseline results

[collecting statistics in a large corpus]

collecting statistics in a large corpus

[evaluation of summarization quality]

evaluation of summarization quality

[FERRET an question-answering Q/A system]

FERRET an question-answering Q/A system

[improvement of 22-38%]

improvement of 22-38%

[argumentation system by Konolige]

argumentation system by Konolige

[evaluation of machine translation]

evaluation of machine translation

[name recognition]

name recognition

[ordering model]

ordering model

[information for use]

information for use

[part of unification-based grammar parsing]

part of unification-based grammar parsing

[speed-up over a SUN]

speed-up over a SUN

[promising features]

promising features

[reflecting the relationship ranks]

reflecting the relationship ranks

[new paradigm for interaction]

new paradigm for interaction

[videodisc]

videodisc

[correlation between the automatic evaluation]

correlation between the automatic evaluation

[respectively on the October test set]

respectively on the October test set

[long phrases]

long phrases

[integral process]

integral process

[protocols from answers]

protocols from answers

[Following recent developments in the evaluation]

Following recent developments in the evaluation

[apartment]

apartment

[edges as in treatments]

edges as in treatments

[meaning through the use]

meaning through the use

[sets of high-precision rules]

sets of high-precision rules

[requiring manual transcription]

requiring manual transcription

[new concepts]

new concepts

[corner parsing algorithm]

corner parsing algorithm

[word features]

word features

[improvement using the SI corpus]

improvement using the SI corpus

[Using a word sense]

Using a word sense

[using a network]

using a network

[99% accuracy]

99% accuracy

[advantages from the point of view]

advantages from the point of view

[robust method for speech understanding]

robust method for speech understanding

[inference method]

inference method

[good insight]

good insight

[one speed-up element]

one speed-up element

[Chinese word sense]

Chinese word sense

[Other contextual clues]

Other contextual clues

[90 ,000 entries]

90 ,000 entries

[question answering capability]

question answering capability

[number of edges]

number of edges

[i Robust parsing of Korean language]

i Robust parsing of Korean language

[only edges]

only edges

[multi-site common evaluation]

multi-site common evaluation

[various ambiguities]

various ambiguities

[word tokens]

word tokens

[system development to domains]

system development to domains

[monitoring the collection]

monitoring the collection

[few years]

few years

[two outputs from information extraction IE]

two outputs from information extraction IE

[model parameters than similar models]

model parameters than similar models

[final language with case markers]

final language with case markers

[involving]

involving

[reflecting the social relationship]

reflecting the social relationship

[candidates for the errors]

candidates for the errors

[identification of 2-character Chinese names]

identification of 2-character Chinese names

[Chinese names]

Chinese names

[various features from the contexts]

various features from the contexts

[inflow of multilingual data]

inflow of multilingual data

[using knowledgebases]

using knowledgebases

[rule-reduction algorithm applying information]

rule-reduction algorithm applying information

[Coalition Language System]

Coalition Language System

[clues]

clues

[extending activity in a manner]

extending activity in a manner

[best hypothesis corresponding]

best hypothesis corresponding

[large-scale acquisition the construction]

large-scale acquisition the construction

[transition probabilities]

transition probabilities

[lists of the feature sets]

lists of the feature sets

[multiple meanings]

multiple meanings

[Task-based evaluation using information retrieval]

Task-based evaluation using information retrieval

[dependency parser]

dependency parser

[satisfying human communication needs]

satisfying human communication needs

[sentence hypotheses]

sentence hypotheses

[indicators for the top-level prediction task]

indicators for the top-level prediction task

[environment TGE]

environment TGE

[rule-reduction algorithm]

rule-reduction algorithm

[novel evaluation method]

novel evaluation method

[35 .0% improvement over system]

35 .0% improvement over system

[much work]

much work

[missing]

missing

[methods LSA-based methods]

methods LSA-based methods

[lack of sense-tagged data]

lack of sense-tagged data

[heuristic learning]

heuristic learning

[QA system performance]

QA system performance

[large list of sentence plans]

large list of sentence plans

[involving automatic extraction]

involving automatic extraction

[likelihood criterion]

likelihood criterion

[distinct domains]

distinct domains

[strategies one utilizing mechanisms]

strategies one utilizing mechanisms

[i parsing]

i parsing

[source dependency]

source dependency

[efficient broad-coverage parser]

efficient broad-coverage parser

[.1% a reduction in error]

.1% a reduction in error

[Language understanding work]

Language understanding work

[identifying patterns in translation data]

identifying patterns in translation data

[use of priors in conditional models]

use of priors in conditional models

[improvement on machine translation tests]

improvement on machine translation tests

[improvements over the WI systems]

improvements over the WI systems

[two systems]

two systems

[state-of-the-art machine translation system]

state-of-the-art machine translation system

[abduction of temporal relations]

abduction of temporal relations

[main parser]

main parser

[individual rules]

individual rules

[method adding annotation]

method adding annotation

[comparing the expressiveness]

comparing the expressiveness

[parsing time complexity]

parsing time complexity

[amount of internal variation]

amount of internal variation

[inflective languages]

inflective languages

[learning of phrase translations]

learning of phrase translations

[strategies utilizing primarily knowledge-based mechanisms]

strategies utilizing primarily knowledge-based mechanisms

[language with overt case markers]

language with overt case markers

[taking advantage]

taking advantage

[major problems including vocabulary]

major problems including vocabulary

[men]

men

[more traditional language interfaces]

more traditional language interfaces

[planning algorithm]

planning algorithm

[attributes such as size perspective-taking]

attributes such as size perspective-taking

[simple explanation]

simple explanation

[reasoning with dispositions]

reasoning with dispositions

[Examination of the effect]

Examination of the effect

[New experimental results]

New experimental results

[using processors APs]

using processors APs

[domain independent acoustic models with classifiers]

domain independent acoustic models with classifiers

[copy of texts]

copy of texts

[i parsing language with case markers]

i parsing language with case markers

[Chinese names without title]

Chinese names without title

[generalization corresponding to arcs]

generalization corresponding to arcs

[evaluation of word sense]

evaluation of word sense

[example discourses]

example discourses

[pronominal anaphora resolution module]

pronominal anaphora resolution module

[more remote events]

more remote events

[sentence-level equivalence classification]

sentence-level equivalence classification

[imitation of performance]

imitation of performance

[broad-coverage parser for English]

broad-coverage parser for English

[ideal]

ideal

[tree adjoining]

tree adjoining

[abduction]

abduction

[men like mostly women]

men like mostly women

[analysis more robust---an agenda-based scheduling parser]

analysis more robust---an agenda-based scheduling parser

[cross-relations between English]

cross-relations between English

[indexation algorithm]

indexation algorithm

[limitations in comparison]

limitations in comparison

[LCS]

LCS

[using conventional word-trigram recognition]

using conventional word-trigram recognition

[pool of WSD systems]

pool of WSD systems

[eliciting human judgments]

eliciting human judgments

[i parsing of Korean language]

i parsing of Korean language

[approach of grammar tailoring]

approach of grammar tailoring

[probabilistic mapping]

probabilistic mapping

[relationship e .g]

relationship e .g

[NE types]

NE types

[target domain]

target domain

[deal of language texts]

deal of language texts

[parser/generator a transfer component]

parser/generator a transfer component

[tutorial dialogue]

tutorial dialogue

[definition questions]

definition questions

[utterances from sites]

utterances from sites

[Korean-to-English machine translation system]

Korean-to-English machine translation system

[109 training speakers]

109 training speakers

[training SPoT on the basis]

training SPoT on the basis

[i parsing a verb final language]

i parsing a verb final language

[word-sense disambiguation systems]

word-sense disambiguation systems

[content understanding]

content understanding

[style and/or]

style and/or

[aspects of the predictive performance]

aspects of the predictive performance

[NLP problems]

NLP problems

[development time]

development time

[dictionaries MRD's]

dictionaries MRD's

[insightful counterexamples]

insightful counterexamples

[pairs of probabilistic grammars]

pairs of probabilistic grammars

[machine translation modules]

machine translation modules

[corpus analysis accessible to researchers]

corpus analysis accessible to researchers

[machine translation quality]

machine translation quality

[progressing from generation]

progressing from generation

[strategies one]

strategies one

[utterances of spontaneous speech]

utterances of spontaneous speech

[learning approach building a stemmer]

learning approach building a stemmer

[middle]

middle

[kinds including reference failures]

kinds including reference failures

[design for agents]

design for agents

[kinds of misunderstandings]

kinds of misunderstandings

[collaborations]

collaborations

[technology of personal computers]

technology of personal computers

[using bilingual parallel corpora a resource]

using bilingual parallel corpora a resource

[different LM]

different LM

[data structure]

data structure

[12,000 utterances from five sites]

12,000 utterances from five sites

[segment contiguity models]

segment contiguity models

[i parsing final language]

i parsing final language

[analysis robust---an agenda-based scheduling parser]

analysis robust---an agenda-based scheduling parser

[conditions for the use]

conditions for the use

[improving the performance]

improving the performance

[free word]

free word

[scoring system output]

scoring system output

[strategies for language understanding]

strategies for language understanding

[industry watch]

industry watch

[small sentences parallel corpus]

small sentences parallel corpus

[information from a large inflow]

information from a large inflow

[Language Processing NLP systems]

Language Processing NLP systems

[acquisition of word-semantic information]

acquisition of word-semantic information

[155 million word]

155 million word

[adjacency graph's Laplacian]

adjacency graph's Laplacian

[correcting homophone errors in compound nouns]

correcting homophone errors in compound nouns

[modelling]

modelling

[deal of natural language texts]

deal of natural language texts

[use in a evaluation]

use in a evaluation

[more frequently predictors]

more frequently predictors

[temporal adverbs]

temporal adverbs

[clustering process]

clustering process

[inherently a conglomeration of individual rules]

inherently a conglomeration of individual rules

[annealing]

annealing

[common claim]

common claim

[Using these ideas]

Using these ideas

[path of arcs]

path of arcs

[geometric model]

geometric model

[weather forecasts]

weather forecasts

[associative processors APs]

associative processors APs

[parsing models]

parsing models

[automatic scheme]

automatic scheme

[last years]

last years

[use in such MT]

use in such MT

[hardness results]

hardness results

[unification algorithms]

unification algorithms

[two algorithms]

two algorithms

[proper substitutions]

proper substitutions

[meaning structure]

meaning structure

[employing a generative approach]

employing a generative approach

[English-Chinese corpora]

English-Chinese corpora

[forward-backward algorithm]

forward-backward algorithm

[more redundancy]

more redundancy

[scope information of the structure]

scope information of the structure

[interactive environment]

interactive environment

[Robust efficient parsing a verb]

Robust efficient parsing a verb

[problem in user modeling]

problem in user modeling

[two datasets]

two datasets

[importance to Machine Translation]

importance to Machine Translation

[Lexical-Functional Grammars]

Lexical-Functional Grammars

[additional 500 features]

additional 500 features

[such linguistic databases]

such linguistic databases

[automatically acquiring new stems]

automatically acquiring new stems

[predicting topic shifts to the problem]

predicting topic shifts to the problem

[using a phrase]

using a phrase

[number of abstracts]

number of abstracts

[complex language processing]

complex language processing

[segment boundaries]

segment boundaries

[holding candidates for understanding results]

holding candidates for understanding results

[entering through message passing]

entering through message passing

[writing]

writing

[89 .75% F-measure a 13% decrease]

89 .75% F-measure a 13% decrease

[potential increase in MRR]

potential increase in MRR

[investigating the combinatorics]

investigating the combinatorics

[language understanding system]

language understanding system

[triplets of words]

triplets of words

[word sense induction]

word sense induction

[using word-trigram recognition]

using word-trigram recognition

[linguistic analysis component]

linguistic analysis component

[simple sentence-plan-generator SPG]

simple sentence-plan-generator SPG

[SimTran]

SimTran

[three dimensions of user models]

three dimensions of user models

[appealing terms]

appealing terms

[precise way]

precise way

[less memory]

less memory

[iii system development via knowledge-based acquisition]

iii system development via knowledge-based acquisition

[relations between entities]

relations between entities

[factors behind this finding including limitations]

factors behind this finding including limitations

[explanation of such discourse phenomena]

explanation of such discourse phenomena

[information from the parse tree]

information from the parse tree

[DARPA speech]

DARPA speech

[understanding technology to language understanding]

understanding technology to language understanding

[histogram]

histogram

[estimation]

estimation

[system development via acquisition]

system development via acquisition

[Following developments in the automatic evaluation]

Following developments in the automatic evaluation

[translation of dependency structures]

translation of dependency structures

[many different forms]

many different forms

[many structures]

many structures

[classes of algorithms]

classes of algorithms

[i efficient parsing of Korean]

i efficient parsing of Korean

[results on addressee identification meetings]

results on addressee identification meetings

[work harvesting bitexts of the laws]

work harvesting bitexts of the laws

[extending the CSR corpus]

extending the CSR corpus

[string with the performance (typically]

string with the performance (typically

[database of TV shows]

database of TV shows

[two word-sense disambiguation systems]

two word-sense disambiguation systems

[Chinese Language Processing]

Chinese Language Processing

[effective use of priors]

effective use of priors

[chain of syllables]

chain of syllables

[commands]

commands

[progressing into the output]

progressing into the output

[parallel sentences in non-parallel corpora]

parallel sentences in non-parallel corpora

[language understanding technology to language understanding]

language understanding technology to language understanding

[nature of the fragments]

nature of the fragments

[speech acts]

speech acts

[development model]

development model

[probabilistic parsing models]

probabilistic parsing models

[Street]

Street

[retrieval accuracy superior]

retrieval accuracy superior

[traditional language interfaces]

traditional language interfaces

[Dynamic Phrasal Lexicon DHPL [Zernik88]]

Dynamic Phrasal Lexicon DHPL [Zernik88]

[recent claims]

recent claims

[requiring development time]

requiring development time

[homophone errors in compound nouns]

homophone errors in compound nouns

[Rapid system development to new domains]

Rapid system development to new domains

[framework restrictive]

framework restrictive

[Hierarchical Phrasal Lexicon]

Hierarchical Phrasal Lexicon

[selection]

selection

[Boolean conditions]

Boolean conditions

[words from high-accuracy alignment models]

words from high-accuracy alignment models

[corpus consisting of ,610 sentences]

corpus consisting of ,610 sentences

[multi-site evaluation of speech]

multi-site evaluation of speech

[forest]

forest

[equivalence between the use]

equivalence between the use

[language model combination]

language model combination

[generation of true text]

generation of true text

[using tag sequences]

using tag sequences

[increase of correlation]

increase of correlation

[using triplets instead of pairs]

using triplets instead of pairs

[collection of utterances of speech]

collection of utterances of speech

[transforming a disposition]

transforming a disposition

[most benefit]

most benefit

[extracting information]

extracting information

[text processing software]

text processing software

[improvement in recognition accuracy]

improvement in recognition accuracy

[comparison with WSD models]

comparison with WSD models

[already in place]

already in place

[men like women]

men like women

[analysis of answers]

analysis of answers

[Learning Model SLM]

Learning Model SLM

[averaging the statistics]

averaging the statistics

[different answering agents searching]

different answering agents searching

[Switchboard dialogues]

Switchboard dialogues

[preceding tag contexts]

preceding tag contexts

[achieving similar performance]

achieving similar performance

[novel solution]

novel solution

[much debate]

much debate

[effect of the sense]

effect of the sense

[verbal means]

verbal means

[accurate word]

accurate word

[average times]

average times

[99% correct sentences]

99% correct sentences

[information piece]

information piece

[.4M English words]

.4M English words

[validity of such linguistic databases]

validity of such linguistic databases

[classification method]

classification method

[state dynamic]

state dynamic

[volume]

volume

[10K sentences parallel corpus]

10K sentences parallel corpus

[principle-based parser for English]

principle-based parser for English

[approach for evaluation]

approach for evaluation

[means heuristic learning]

means heuristic learning

[automatic translation]

automatic translation

[broad use including jointly conditioning]

broad use including jointly conditioning

[effective use of priors in models]

effective use of priors in models

[paraphrasing of output]

paraphrasing of output

[breakdown]

breakdown

[user evaluation]

user evaluation

[model parameters than phrase-based models]

model parameters than phrase-based models

[using a non-terminal symbol]

using a non-terminal symbol

[abduction of relations]

abduction of relations

[Language Processing NLP]

Language Processing NLP

[object-oriented language parsing]

object-oriented language parsing

[utilizing examples]

utilizing examples

[NTHU's system]

NTHU's system

[coverage lexicon]

coverage lexicon

[involving extraction of translation lexicons]

involving extraction of translation lexicons

[generative approach]

generative approach

[two disambiguation systems one]

two disambiguation systems one

[ii High quality translation]

ii High quality translation

[automating certain aspects]

automating certain aspects

[also in the areas]

also in the areas

[arcs between nodes]

arcs between nodes

[evaluation results involving extraction from text]

evaluation results involving extraction from text

[category transitions]

category transitions

[Arabic's morphology]

Arabic's morphology

[corpus of the language]

corpus of the language

[existing parsing models]

existing parsing models

[unsupervised technique]

unsupervised technique

[eliciting judgments]

eliciting judgments

[event-based summarization approaches]

event-based summarization approaches

[expressions in English]

expressions in English

[disambiguation the interpretation]

disambiguation the interpretation

[.4M words]

.4M words

[formulas with values]

formulas with values

[judges]

judges

[Applications of path-based inference rules]

Applications of path-based inference rules

[building polarity-tagged corpus]

building polarity-tagged corpus

[45% reduction]

45% reduction

[155 million word corpus]

155 million word corpus

[resulting word segmentation system]

resulting word segmentation system

[relevance]

relevance

[important bearing on reasoning]

important bearing on reasoning

[concerning translation equivalence]

concerning translation equivalence

[scheme collecting statistics]

scheme collecting statistics

[real-valued features]

real-valued features

[subcategorization frame distributions]

subcategorization frame distributions

[results casting doubt]

results casting doubt

[acquisition of information e]

acquisition of information e

[two formalisms Tree Grammars]

two formalisms Tree Grammars

[key factors]

key factors

[light of functions]

light of functions

[WSD evaluation]

WSD evaluation

[aspects including the influence]

aspects including the influence

[understanding system reasoning]

understanding system reasoning

[acquisition of constraints in broad domains]

acquisition of constraints in broad domains

[Most text-understanding systems]

Most text-understanding systems

[translation lexicons]

translation lexicons

[phrase boundary]

phrase boundary

[constraints on the translation relation]

constraints on the translation relation

[data relevant to task]

data relevant to task

[inducing verb classes from corpus data]

inducing verb classes from corpus data

[wide coverage grammar]

wide coverage grammar

[using dialogue data]

using dialogue data

[using an out-of-domain corpus]

using an out-of-domain corpus

[75% accuracy]

75% accuracy

[reestimation]

reestimation

[iv modeling of word features]

iv modeling of word features

[pre-existing components]

pre-existing components

[important role in cognition]

important role in cognition

[proposition with implicit quantifiers]

proposition with implicit quantifiers

[semantic roles]

semantic roles

[coordinate sentences]

coordinate sentences

[planning-based architecture]

planning-based architecture

[applying automated evaluation techniques]

applying automated evaluation techniques

[explanation for the nature]

explanation for the nature

[three methods methods]

three methods methods

[application of Unification Grammar UCG]

application of Unification Grammar UCG

[Wall Street Journal treebank]

Wall Street Journal treebank

[results casting doubt on this assumption]

results casting doubt on this assumption

[addition to an component]

addition to an component

[numbering]

numbering

[domain of HK laws]

domain of HK laws

[features of a representation]

features of a representation

[Cooperative Model for language understanding]

Cooperative Model for language understanding

[more data]

more data

[corpus resource]

corpus resource

[representation the Phrasal Lexicon]

representation the Phrasal Lexicon

[properties of grammars]

properties of grammars

[directing the process]

directing the process

[corresponding with the confidence]

corresponding with the confidence

[such databases]

such databases

[methods constructing word vectors]

methods constructing word vectors

[Several advantages from the point]

Several advantages from the point

[basis for the large-scale acquisition]

basis for the large-scale acquisition

[grammar parsing]

grammar parsing

[neat relation]

neat relation

[output of an parser]

output of an parser

[fulfillment of well-formedness conditions]

fulfillment of well-formedness conditions

[precision over unstemmed text]

precision over unstemmed text

[techniques for analysis]

techniques for analysis

[parsing time]

parsing time

[generative character recognition OCR model]

generative character recognition OCR model

[general preference]

general preference

[possible word-senses]

possible word-senses

[formalization]

formalization

[SMT system]

SMT system

[domain independent models with off-the-shelf classifiers]

domain independent models with off-the-shelf classifiers

[achieving baseline results in a time]

achieving baseline results in a time

[understanding system with intelligent agents]

understanding system with intelligent agents

[language modeling of speech]

language modeling of speech

[meeting context]

meeting context

[bilingual resource]

bilingual resource

[machine translation system alone]

machine translation system alone

[fully-connected class grammar]

fully-connected class grammar

[combination of structures]

combination of structures

[domain entities]

domain entities

[domain strategy for the articulation]

domain strategy for the articulation

[namely the algorithm]

namely the algorithm

[grammar generation system]

grammar generation system

[Konolige]

Konolige

[various kinds including reference failures]

various kinds including reference failures

[event type]

event type

[categories in trees]

categories in trees

[manual evaluation]

manual evaluation

[other strategies]

other strategies

[employing n-gram models for prediction]

employing n-gram models for prediction

[only 14 .0%]

only 14 .0%

[word-sense disambiguation systems one]

word-sense disambiguation systems one

[evaluation of strings]

evaluation of strings

[three domain-independent capabilities]

three domain-independent capabilities

[vagueness of the concepts]

vagueness of the concepts

[SI recognition]

SI recognition

[two language pairs]

two language pairs

[advantage of the patterns]

advantage of the patterns

[Statistical Learning Model]

Statistical Learning Model

[direct imitation]

direct imitation

[correcting these errors in bunsetsu]

correcting these errors in bunsetsu

[computational algorithms]

computational algorithms

[word co-occurrences]

word co-occurrences

[different strategies one]

different strategies one

[computation of the form]

computation of the form

[document classification]

document classification

[ambiguity packing techniques]

ambiguity packing techniques

[culling information]

culling information

[information e]

information e

[recent work harvesting bitexts]

recent work harvesting bitexts

[translingual reach]

translingual reach

[improvement for speaker adaptation]

improvement for speaker adaptation

[designing with the Senseval series]

designing with the Senseval series

[automatically gathering a number]

automatically gathering a number

[automatic evaluation method]

automatic evaluation method

[knowledge about dialogs]

knowledge about dialogs

[picking the best hypothesis]

picking the best hypothesis

[pairs of context-free grammars working]

pairs of context-free grammars working

[system with agents]

system with agents

[individually good predictors]

individually good predictors

[problem combining language models]

problem combining language models

[such languages]

such languages

[gathering a number]

gathering a number

[simple interpolation methods like interpolation]

simple interpolation methods like interpolation

[Various properties]

Various properties

[providing an interpretation]

providing an interpretation

[decrease]

decrease

[third of the sentences]

third of the sentences

[parsing of Korean]

parsing of Korean

[categories of the words]

categories of the words

[new linguistic representation]

new linguistic representation

[suppressing the fuzzy quantifier]

suppressing the fuzzy quantifier

[making syntactic analysis robust---an scheduling parser]

making syntactic analysis robust---an scheduling parser

[argumentation system]

argumentation system

[Non]

Non

[scalability]

scalability

[combiner with hard decisions]

combiner with hard decisions

[sophisticated representations for document classification]

sophisticated representations for document classification

[large amount]

large amount

[new paradigm for SI training]

new paradigm for SI training

[reduction in the search space]

reduction in the search space

[interpolation methods]

interpolation methods

[two new disambiguation systems]

two new disambiguation systems

[very small corpus]

very small corpus

[paraphrases meaning equivalence]

paraphrases meaning equivalence

[such as systems]

such as systems

[parsing a verb]

parsing a verb

[appealing terms of simplicity]

appealing terms of simplicity

[comparing the expressiveness of the formalisms]

comparing the expressiveness of the formalisms

[antecedence recovery]

antecedence recovery

[subset 14 .0%]

subset 14 .0%

[other kernel functions]

other kernel functions

[word order generation]

word order generation

[San Diego]

San Diego

[set of phrases]

set of phrases

[.24%]

.24%

[generation of text]

generation of text

[probabilistic optical character recognition model]

probabilistic optical character recognition model

[more available resource]

more available resource

[machine learning experiments]

machine learning experiments

[terms of accuracy]

terms of accuracy

[Lucy a understanding system]

Lucy a understanding system

[including a word]

including a word

[Hong Kong]

Hong Kong

[Using only 40 utterances for adaptation]

Using only 40 utterances for adaptation

[type of finite automaton]

type of finite automaton

[stages of 1]

stages of 1

[task of semantic interpretation]

task of semantic interpretation

[research on major problems including vocabulary]

research on major problems including vocabulary

[many grammar formalisms]

many grammar formalisms

[IBM statistical MT system]

IBM statistical MT system

[text hierarchy]

text hierarchy

[therefore a choice for the parser]

therefore a choice for the parser

[harvesting bitexts of the laws]

harvesting bitexts of the laws

[extractive mutli-document summarization approaches]

extractive mutli-document summarization approaches

[definition of typicality]

definition of typicality

[parsing language]

parsing language

[State-of-the-art Question]

State-of-the-art Question

[Resource Management RM corpus]

Resource Management RM corpus

[width]

width

[hypertext links]

hypertext links

[culling information from a inflow]

culling information from a inflow

[algorithm automatically acquiring stems]

algorithm automatically acquiring stems

[indicators]

indicators

[message passing between actors]

message passing between actors

[broad-coverage principle-based parser for English]

broad-coverage principle-based parser for English

[role from the discourse]

role from the discourse

[techniques making syntactic analysis]

techniques making syntactic analysis

[corpus of 110 words]

corpus of 110 words

[Using a Chinese word sense]

Using a Chinese word sense

[large number of abstracts]

large number of abstracts

[theory of plausible reasoning]

theory of plausible reasoning

[principle advantage]

principle advantage

[head-to-head tests]

head-to-head tests

[employing a traditional phonological approach]

employing a traditional phonological approach

[message passing]

message passing

[linking to translation equivalents]

linking to translation equivalents

[N-gram]

N-gram

[state of research]

state of research

[markers]

markers

[extractive summarization approaches]

extractive summarization approaches

[yielding ways]

yielding ways

[word alignment component]

word alignment component

[speech dialogue interactions]

speech dialogue interactions

[field exercises with the Marines]

field exercises with the Marines

[language model of moves]

language model of moves

[outstanding problems]

outstanding problems

[subject of much debate]

subject of much debate

[parallel]

parallel

[focusing on the changes]

focusing on the changes

[literal meaning of input]

literal meaning of input

[such as size b perspective-taking]

such as size b perspective-taking

[demonstration of an Language System]

demonstration of an Language System

[constraints on the attributes]

constraints on the attributes

[shallow features of questions]

shallow features of questions

[7 .5% word]

7 .5% word

[i parsing of Korean]

i parsing of Korean

[ellipsis]

ellipsis

[Transfer-Driven Machine Translation]

Transfer-Driven Machine Translation

[structural disambiguation]

structural disambiguation

[robust recognition]

robust recognition

[inducing a grammar from corpora]

inducing a grammar from corpora

[a attributes such as size perspective-taking]

a attributes such as size perspective-taking

[use of priors in loglinear models]

use of priors in loglinear models

[iv fine-grained modeling of word features]

iv fine-grained modeling of word features

[text collection covering the domain]

text collection covering the domain

[Entity]

Entity

[tests against one]

tests against one

[spelling-checkers for English]

spelling-checkers for English

[phonological knowledge base system]

phonological knowledge base system

[different language system domains]

different language system domains

[warfare]

warfare

[empirical test]

empirical test

[advantage of information]

advantage of information

[incorporating an approximation]

incorporating an approximation

[traditional generative phonological approach]

traditional generative phonological approach

[class NP]

class NP

[paradigm for human interaction]

paradigm for human interaction

[frame semantics]

frame semantics

[corpora a more available resource]

corpora a more available resource

[non-English Arabic stemmer]

non-English Arabic stemmer

[integrating Q/A applications]

integrating Q/A applications

[part of the processing time]

part of the processing time

[QA system]

QA system

[novel automatic evaluation method]

novel automatic evaluation method

[starting with a corpus words]

starting with a corpus words

[Schutze's idea]

Schutze's idea

[parsing method]

parsing method

[unsupervised algorithm acquiring new stems]

unsupervised algorithm acquiring new stems

[reflecting the relationship e .g ranks]

reflecting the relationship e .g ranks

[traditional phonological approach]

traditional phonological approach

[exploring bitexts in a volume]

exploring bitexts in a volume

[areas during IDUS development]

areas during IDUS development

[grammaticality at least 99% correct sentences]

grammaticality at least 99% correct sentences

[interpolation of N-grams]

interpolation of N-grams

[45% reduction in error]

45% reduction in error

[efficient parsing language]

efficient parsing language

[importance to Statistical Machine Translation]

importance to Statistical Machine Translation

[using the SemCor data sets]

using the SemCor data sets

[natural language search engines]

natural language search engines

[logical well-formedness conditions]

logical well-formedness conditions

[MUC-3]

MUC-3

[principle-based parser]

principle-based parser

[also graphs]

also graphs

[use of representations]

use of representations

[domain independent models with classifiers]

domain independent models with classifiers

[MLE reestimation algorithms the algorithm]

MLE reestimation algorithms the algorithm

[different aspects]

different aspects

[study of anaphoric expressions]

study of anaphoric expressions

[adopting a simple paraphrasing technique]

adopting a simple paraphrasing technique

[role in cognition]

role in cognition

[Paramax language]

Paramax language

[efficiency—to]

efficiency—to

[interactive systems]

interactive systems

[antecedence of each element]

antecedence of each element

[natural settings]

natural settings

[inducing verb classes]

inducing verb classes

[only .0%]

only .0%

[grammar formalisms such as rewriting systems]

grammar formalisms such as rewriting systems

[sentences in comparable corpora]

sentences in comparable corpora

[research articles]

research articles

[language model score]

language model score

[shallow features]

shallow features

[three-tiered approach]

three-tiered approach

[decision tree building]

decision tree building

[information extraction IE]

information extraction IE

[previous papers]

previous papers

[automatic generation]

automatic generation

[optical character recognition OCR]

optical character recognition OCR

[poor construction]

poor construction

[other indices]

other indices

[alignment models]

alignment models

[efficient broad-coverage parser for English]

efficient broad-coverage parser for English

[dynamic combiner with decisions]

dynamic combiner with decisions

[management]

management

[Intelligent Document Understanding System]

Intelligent Document Understanding System

[way identifying predicate argument structures]

way identifying predicate argument structures

[exploiting sophisticated representations]

exploiting sophisticated representations

[generation of hypertext links]

generation of hypertext links

[speech data]

speech data

[good choice for the parser]

good choice for the parser

[many grammatical formalisms]

many grammatical formalisms

[entity-oriented approach to parsing]

entity-oriented approach to parsing

[fundamentally different strategies one]

fundamentally different strategies one

[including conditioning on consecutive words]

including conditioning on consecutive words

[Chomsky's]

Chomsky's

[papers [Zernik87]]

papers [Zernik87]

[ambiguity packing stochastic disambiguation techniques]

ambiguity packing stochastic disambiguation techniques

[Robust parsing final language]

Robust parsing final language

[n-best speech/language integration architecture]

n-best speech/language integration architecture

[scope information of the syntactic structure]

scope information of the syntactic structure

[means for grounding]

means for grounding

[acquisition of semantic constraints]

acquisition of semantic constraints

[such redundancy]

such redundancy

[edges with a interpretation]

edges with a interpretation

[speakers prior to training]

speakers prior to training

[translation data]

translation data

[phrase-based machine translation method]

phrase-based machine translation method

[node structures]

node structures

[increase in MRR]

increase in MRR

[bilingual material]

bilingual material

[using a speech from speakers]

using a speech from speakers

[algorithm for phrase structure parsing]

algorithm for phrase structure parsing

[document preference]

document preference

[most young men like women]

most young men like women

[respect to the question]

respect to the question

[internal variation in a set]

internal variation in a set

[Finite State Model]

Finite State Model

[introspection in the hope]

introspection in the hope

[set of model parameters]

set of model parameters

[efficient principle-based parser]

efficient principle-based parser

[acquisition of semantic constraints in domains]

acquisition of semantic constraints in domains

[Listen-Communicate-Show]

Listen-Communicate-Show

[i efficient parsing a verb]

i efficient parsing a verb

[weighting]

weighting

[14 .0%]

14 .0%

[using training speakers]

using training speakers

[Coalition]

Coalition

[i Robust parsing final language]

i Robust parsing final language

[word sense induction WSI]

word sense induction WSI

[consisting of sentences]

consisting of sentences

[addressee identification]

addressee identification

[Interactive Machine Translation]

Interactive Machine Translation

[evaluation algorithms]

evaluation algorithms

[using semantic syntactic categories]

using semantic syntactic categories

[making corpus analysis]

making corpus analysis

[model in place]

model in place

[MT evaluation methods (BLEU]

MT evaluation methods (BLEU

[design of the discourse understanding process]

design of the discourse understanding process

[sentence-plan-ranker]

sentence-plan-ranker

[formalisms Tree Grammars]

formalisms Tree Grammars

[major new problems]

major new problems

[utilizing mechanisms]

utilizing mechanisms

[mutual information]

mutual information

[story link]

story link

[unsupervised algorithm automatically acquiring stems]

unsupervised algorithm automatically acquiring stems

[two problems]

two problems

[various kinds]

various kinds

[token classification]

token classification

[discovering sentences in comparable corpora]

discovering sentences in comparable corpora

[reasoning in machine translation]

reasoning in machine translation

[parallel corpus 100 words]

parallel corpus 100 words

[new part-of-speech tagger explicit use]

new part-of-speech tagger explicit use

[starting with a corpus]

starting with a corpus

[such as cue phrases]

such as cue phrases

[various training factors on performance]

various training factors on performance

[two new word-sense systems]

two new word-sense systems

[term weighting]

term weighting

[best tagging result]

best tagging result

[proposals]

proposals

[sentence-aligned corpus]

sentence-aligned corpus

[various functions]

various functions

[least 96% paraphrases]

least 96% paraphrases

[achieving baseline results]

achieving baseline results

[meetings using Bayesian Network classifiers]

meetings using Bayesian Network classifiers

[probable morpheme sequence]

probable morpheme sequence

[relationship e .g social ranks]

relationship e .g social ranks

[Manual acquisition of constraints in domains]

Manual acquisition of constraints in domains

[analyzing metaphors]

analyzing metaphors

[light of aspects]

light of aspects

[independent extractive summarization approaches]

independent extractive summarization approaches

[generalization roughly corresponding to arcs]

generalization roughly corresponding to arcs

[Japanese sentence analyses]

Japanese sentence analyses

[predicting topic shifts]

predicting topic shifts

[applying information]

applying information

[including conditioning on multiple words]

including conditioning on multiple words

[probabilistic model of natural language]

probabilistic model of natural language

[hand-crafted generation component]

hand-crafted generation component

[adopting techniques]

adopting techniques

[posteriori estimation of Markov models]

posteriori estimation of Markov models

[character recognition OCR]

character recognition OCR

[various features]

various features

[spatial descriptions]

spatial descriptions

[Thai-English language]

Thai-English language

[several factors including limitations]

several factors including limitations

[thus the number]

thus the number

[performance vs]

performance vs

[common assumption]

common assumption

[also graphs indispensable]

also graphs indispensable

[algorithms for the analysis]

algorithms for the analysis

[qualitative spatial constraints]

qualitative spatial constraints

[best previous tagging result]

best previous tagging result

[noisy output]

noisy output

[correlation between evaluation measures]

correlation between evaluation measures

[using a little speech]

using a little speech

[mission success]

mission success

[several kinds]

several kinds

[Manual acquisition of semantic constraints]

Manual acquisition of semantic constraints

[classifying SCF data]

classifying SCF data

[several levels]

several levels

[mathematical formalism]

mathematical formalism

[features of the system]

features of the system

[graphs indispensable]

graphs indispensable

[methods constructing word vectors .e methods]

methods constructing word vectors .e methods

[formalization of reasoning]

formalization of reasoning

[reformulation of the sentence]

reformulation of the sentence

[meaning of an utterance]

meaning of an utterance

[syntax formalisms]

syntax formalisms

[support]

support

[automatic parameter optimization]

automatic parameter optimization

[using the reference]

using the reference

[Dependency Diagram CDD]

Dependency Diagram CDD

[discourse phenomena as cue phrases]

discourse phenomena as cue phrases

[mixtures of techniques]

mixtures of techniques

[ranking of these parses]

ranking of these parses

[forms in the input documents]

forms in the input documents

[two outputs from information extraction]

two outputs from information extraction

[sentences corpus]

sentences corpus

[using associative processors]

using associative processors

[type of automaton]

type of automaton

[framework less restrictive than ones]

framework less restrictive than ones

[96% paraphrases]

96% paraphrases

[parameter optimization]

parameter optimization

[word-semantic information .g]

word-semantic information .g

[nature of these sentences]

nature of these sentences

[heuristic learning of phrase translations]

heuristic learning of phrase translations

[expressions of languages]

expressions of languages

[understanding process]

understanding process

[methods .e methods]

methods .e methods

[traditional natural language interfaces]

traditional natural language interfaces

[example-retrieval]

example-retrieval

[around 97% match accuracy]

around 97% match accuracy

[Resource Management RM]

Resource Management RM

[using features]

using features

[connection]

connection

[drastic generalization of chart Parsing]

drastic generalization of chart Parsing

[reflecting the relationship]

reflecting the relationship

[researchers]

researchers

[FERRET an interactive question-answering system]

FERRET an interactive question-answering system

[most men like mostly women]

most men like mostly women

[part-of-speech tagger explicit use]

part-of-speech tagger explicit use

[combining the results]

combining the results

[use of expressions]

use of expressions

[parsing of Korean a verb]

parsing of Korean a verb

[beam-search decoder]

beam-search decoder

[efficiency of RCL parsers]

efficiency of RCL parsers

[using machine translation output]

using machine translation output

[terms of coherence]

terms of coherence

[user's goals]

user's goals

[research on problems including unlimited vocabulary]

research on problems including unlimited vocabulary

[containing sets]

containing sets

[transfer system]

transfer system

[performance thresholds]

performance thresholds

[new representation the Phrasal Lexicon]

new representation the Phrasal Lexicon

[LHIP a system]

LHIP a system

[system with intelligent agents]

system with intelligent agents

[referential properties]

referential properties

[subgraphs]

subgraphs

[customizable IE paradigm]

customizable IE paradigm

[Owing]

Owing

[determining the vehicle]

determining the vehicle

[Data]

Data

[performance in real applications]

performance in real applications

[a numeric-valued attributes such as size]

a numeric-valued attributes such as size

[behavior of a grammar]

behavior of a grammar

[translation probabilities]

translation probabilities

[determining the literal meaning]

determining the literal meaning

[small sentences corpus]

small sentences corpus

[central problem]

central problem

[exploiting representations]

exploiting representations

[predicting new concepts]

predicting new concepts

[parameter optimization of the WSI algorithm]

parameter optimization of the WSI algorithm

[Significant improvement over word alignment techniques]

Significant improvement over word alignment techniques

[strength of antecedence recovery]

strength of antecedence recovery

[mixture of trigram models]

mixture of trigram models

[variation in a set]

variation in a set

[using criterion]

using criterion

[recent speculative claims]

recent speculative claims

[ground]

ground

[phonological system]

phonological system

[using triplets]

using triplets

[2-character names]

2-character names

[word blocks]

word blocks

[corpus consisting of sentences]

corpus consisting of sentences

[accomplishments of MADCOW]

accomplishments of MADCOW

[linguistic expressiveness of the formalisms]

linguistic expressiveness of the formalisms

[m-th order]

m-th order

[numerous non-literal aspects of communication]

numerous non-literal aspects of communication

[maximally plausible interpretation]

maximally plausible interpretation

[metaphor mappings]

metaphor mappings

[grammaticality correct sentences]

grammaticality correct sentences

[frequently predictors]

frequently predictors

[experimental results on all applications]

experimental results on all applications

[discourse task]

discourse task

[associative processors]

associative processors

[naturally sounding dialog systems]

naturally sounding dialog systems

[pairs of probabilistic grammars working]

pairs of probabilistic grammars working

[island-based parsing method]

island-based parsing method

[cognition]

cognition

[subset only .0%]

subset only .0%

[number of experiments]

number of experiments

[small amount]

small amount

[Boolean]

Boolean

[Offering advantages]

Offering advantages

[SUN 4]

SUN 4

[m-component mixture]

m-component mixture

[likely answer candidates]

likely answer candidates

[user modeling component]

user modeling component

[corpus ,000 words]

corpus ,000 words

[conventional algorithms]

conventional algorithms

[six extracts]

six extracts

[references of the pronoun]

references of the pronoun

[layout structures]

layout structures

[TDMT Transfer-Driven Machine Translation]

TDMT Transfer-Driven Machine Translation

[new SI corpus]

new SI corpus

[automatically gathering a number of abstracts]

automatically gathering a number of abstracts

[properties of tree-adjoining grammars TAG]

properties of tree-adjoining grammars TAG

[constraints in domains]

constraints in domains

[different forms]

different forms

[speaker-independent SI training]

speaker-independent SI training

[word-sense systems one]

word-sense systems one

[rewriting]

rewriting

[perspective-taking in reference]

perspective-taking in reference

[understanding accuracy]

understanding accuracy

[morpheme sequence]

morpheme sequence

[iii Rapid system development to domains]

iii Rapid system development to domains

[corresponding to the LM]

corresponding to the LM

[semantic representation]

semantic representation

[developing the grammars]

developing the grammars

[precision over text]

precision over text

[state-of-the-art evaluation measures]

state-of-the-art evaluation measures

[different strategies utilizing primarily knowledge-based mechanisms]

different strategies utilizing primarily knowledge-based mechanisms

[taking runtime]

taking runtime

[several key factors behind this finding]

several key factors behind this finding

[Natural Language Processing systems]

Natural Language Processing systems

[consecutive words]

consecutive words

[broad use of lexical features]

broad use of lexical features

[vocabulary size]

vocabulary size

[combiner]

combiner

[Applications of inference rules]

Applications of inference rules

[using the language model]

using the language model

[easy exploitation of domain semantics]

easy exploitation of domain semantics

[Using 40 utterances]

Using 40 utterances

[automatically evaluating answers]

automatically evaluating answers

[patent claim]

patent claim

[entering through message passing between actors]

entering through message passing between actors

[7% improvement]

7% improvement

[subcategorization frame SCF distributions]

subcategorization frame SCF distributions

[undisambiguated SCF data]

undisambiguated SCF data

[different answering agents]

different answering agents

[user modeling systems]

user modeling systems

[structure finding methods]

structure finding methods

[time decoding algorithm]

time decoding algorithm

[inflow of multimedia data]

inflow of multimedia data

[system reasoning]

system reasoning

[advantage of the symmetric patterns]

advantage of the symmetric patterns

[English coordinate sentences]

English coordinate sentences

[grammaticality 99% sentences equivalence]

grammaticality 99% sentences equivalence

[pronominal anaphora resolution]

pronominal anaphora resolution

[careful examination of the complexity]

careful examination of the complexity

[Implications automating aspects of language learning]

Implications automating aspects of language learning

[Current language interfaces]

Current language interfaces

[one superiority]

one superiority

[aspects such as communication procedures]

aspects such as communication procedures

[using an argumentation system]

using an argumentation system

[natural language understanding]

natural language understanding

[containing sets of features]

containing sets of features

[ONTOSCORE]

ONTOSCORE

[unmodified subgraphs]

unmodified subgraphs

[picking the hypothesis corresponding]

picking the hypothesis corresponding

[great deal]

great deal

[cooperative responses adaptive]

cooperative responses adaptive

[real-time response]

real-time response

[contributing to the word]

contributing to the word

[New results on all applications]

New results on all applications

[efficient parsing a verb final language]

efficient parsing a verb final language

[WSD systems]

WSD systems

[approach to Web-based computer-assisted writing]

approach to Web-based computer-assisted writing

[information in memory]

information in memory

[edges with a semantic interpretation]

edges with a semantic interpretation

[OCR accuracy]

OCR accuracy

[Marines]

Marines

[contrast to the approach]

contrast to the approach

[argument structures]

argument structures

[using memory]

using memory

[Hierarchical Acyclic Graph HDAG Kernel]

Hierarchical Acyclic Graph HDAG Kernel

[expensive part]

expensive part

[finding data relevant]

finding data relevant

[periods]

periods

[combining language models]

combining language models

[unique properties of tree-adjoining grammars]

unique properties of tree-adjoining grammars

[restrictions]

restrictions

[intelligibility]

intelligibility

[mixed-initiative speech dialogue interactions]

mixed-initiative speech dialogue interactions

[performing inference in semantic networks]

performing inference in semantic networks

[novel Model for language understanding]

novel Model for language understanding

[existing metrics]

existing metrics

[empirical comparison]

empirical comparison

[feature sets]

feature sets

[simple interpolation methods]

simple interpolation methods

[searching for answers in corpora]

searching for answers in corpora

[phrase structure parsing of language]

phrase structure parsing of language

[system development to domains via acquisition]

system development to domains via acquisition

[biasing]

biasing

[determining the literal meaning of input]

determining the literal meaning of input

[type of transfer system]

type of transfer system

[three words from alignment models]

three words from alignment models

[basis for a tool]

basis for a tool

[machine-readable dictionaries]

machine-readable dictionaries

[possible word-senses of words]

possible word-senses of words

[500 features]

500 features

[simpler set of model parameters]

simpler set of model parameters

[new performance enhancing techniques]

new performance enhancing techniques

[theory for natural language presuppositions]

theory for natural language presuppositions

[i Robust efficient parsing]

i Robust efficient parsing

[formalisms Tree Adjoining Grammars]

formalisms Tree Adjoining Grammars

[judgment]

judgment, judgments

[aspects of the interaction requirements]

aspects of the interaction requirements

[system response]

system response

[suppressing the quantifier most]

suppressing the quantifier most

[rarely in the context]

rarely in the context

[Syntax-based statistical machine translation]

Syntax-based statistical machine translation, syntax-based statistical machine translation

[use evaluating the summarization quality]

use evaluating the summarization quality

[discovering sentences in comparable non-parallel corpora]

discovering sentences in comparable non-parallel corpora

[speed-up over a SUN 4]

speed-up over a SUN 4

[identification of names without title]

identification of names without title

[Robust parsing a verb]

Robust parsing a verb

[substring]

substring

[interesting combination of hand-built resources]

interesting combination of hand-built resources

[semantic verb classes]

semantic verb classes

[quality translation via word sense]

quality translation via word sense

[accomplishments monitoring the collection]

accomplishments monitoring the collection

[bootstrapping approach to Entity NE]

bootstrapping approach to Entity NE

[validity of complex linguistic databases]

validity of complex linguistic databases

[new tagger explicit use]

new tagger explicit use

[two contributions to vocabulary]

two contributions to vocabulary

[model for output selection]

model for output selection

[evaluation scheme]

evaluation scheme

[mobile agent]

mobile agents, mobile agent

[advances in Automatic Speech Recognition technology]

advances in Automatic Speech Recognition technology

[impact on models]

impact on models

[major problems]

major problems

[applying evaluation techniques]

applying evaluation techniques

[making analysis robust---an scheduling parser]

making analysis robust---an scheduling parser

[generalization roughly corresponding]

generalization roughly corresponding

[transfer phase]

transfer phase

[probabilistic model of language]

probabilistic model of language

[major problems including unlimited vocabulary]

major problems including unlimited vocabulary

[96% correct paraphrases meaning equivalence]

96% correct paraphrases meaning equivalence

[lexica]

lexica

[covering the specific domain]

covering the specific domain

[small parallel corpus]

small parallel corpus

[results of a evaluation]

results of a evaluation

[processors]

processors

[factors including limitations]

factors including limitations

[using these models in combination]

using these models in combination

[disjoint phrases]

disjoint phrases

[mutli-document summarization approaches]

mutli-document summarization approaches

[paraphrasing technique]

paraphrasing technique

[word order]

word order

[styles performing inference in semantic networks]

styles performing inference in semantic networks

[collection of utterances]

collection of utterances

[phrasal SMT]

phrasal SMT

[novel method discovering sentences]

novel method discovering sentences

[careful examination]

careful examination

[correspondences]

correspondences

[comparable corpora]

comparable corpora

[dimensionality space]

dimensionality space

[three methods LSA-based methods]

three methods LSA-based methods

[starting point]

starting point

[sense of much]

sense of much

[basis for a disambiguation tool]

basis for a disambiguation tool

[so many structures]

so many structures

[context languages]

context languages

[using processors for real-time language translation]

using processors for real-time language translation

[expert systems]

expert systems

[bilingual parallel corpora a resource]

bilingual parallel corpora a resource

[validity of linguistic databases]

validity of linguistic databases

[such case-based MT]

such case-based MT

[domain acoustic models]

domain acoustic models

[dimensions skill level to the system]

dimensions skill level to the system

[SENSEVAL-2 sample task]

SENSEVAL-2 sample task

[exact match]

exact match

[problems with the approach]

problems with the approach

[bootstrapping approach to Entity]

bootstrapping approach to Entity

[ranks of the referents]

ranks of the referents

[results along with a analysis]

results along with a analysis

[entropy classifier]

entropy classifier

[multi engine machine]

multi engine machine

[parsing final language with case markers]

parsing final language with case markers

[English phrases]

English phrases

[.1%]

.1%

[correlation between the evaluation]

correlation between the evaluation

[face-to-face meetings]

face-to-face meetings

[set of tasks]

set of tasks

[drawing a space]

drawing a space

[biasing parsing]

biasing parsing

[integral process in human understanding]

integral process in human understanding

[parsing flexibilities]

parsing flexibilities

[overview]

overview

[language environment]

language environment

[over a third]

over a third

[relational database]

relational database

[evaluation of this method]

evaluation of this method

[corresponding in a chart]

corresponding in a chart

[accuracy superior to any]

accuracy superior to any

[spanning edges]

spanning edges

[domain-independent capabilities]

domain-independent capabilities

[parse forests]

parse forests

[-speaking analyst]

-speaking analyst

[.3%]

.3%

[evaluating answers]

evaluating answers

[common evaluation of speech]

common evaluation of speech

[probabilistic parser]

probabilistic parser

[indexing according to character bigrams]

indexing according to character bigrams

[direct application of metrics]

direct application of metrics

[verbal interactions of dialogs]

verbal interactions of dialogs

[valuable methods of language interfaces]

valuable methods of language interfaces

[hypothesis corresponding with the confidence]

hypothesis corresponding with the confidence

[instead of the practice]

instead of the practice

[parsing a set]

parsing a set

[automatically training]

automatically training

[starting with a corpus ,000 words]

starting with a corpus ,000 words

[high performance in applications]

high performance in applications

[personal names]

personal names

[two new word-sense systems one]

two new word-sense systems one

[performing inference in networks]

performing inference in networks

[improving system performance]

improving system performance

[phonological approach]

phonological approach

[.3M]

.3M

[novel Cooperative Model]

novel Cooperative Model

[better indicators]

better indicators

[complicated large knowledgebases]

complicated large knowledgebases

[worse coverage]

worse coverage

[knowledge about dialog schemata]

knowledge about dialog schemata

[syntactic analysis more robust---an scheduling parser]

syntactic analysis more robust---an scheduling parser

[fact transition graphs]

fact transition graphs

[approach via the introspection]

approach via the introspection

[discourse effect]

discourse effect

[set of word alignments]

set of word alignments

[lexical cohesion-based approach]

lexical cohesion-based approach

[Several extensions of this idea]

Several extensions of this idea

[.4M]

.4M

[behavioral specification]

behavioral specification

[Other clues such as editing terms]

Other clues such as editing terms

[recall in the task]

recall in the task

[new tagger]

new tagger

[using an word alignment]

using an word alignment

[large inflow]

large inflow

[usual introspection]

usual introspection

[Answering]

Answering

[combinatorics of indexation]

combinatorics of indexation

[phrase re-ordering]

phrase re-ordering

[learning methodology applicable]

learning methodology applicable

[method adding annotation to corpora]

method adding annotation to corpora

[Statistical Learning]

Statistical Learning

[examples similar to an input expression]

examples similar to an input expression

[direct application]

direct application

[theory for language presuppositions 1988/]

theory for language presuppositions 1988/

[format for meaning]

format for meaning

[segments of actual descriptions]

segments of actual descriptions

[typical kinds]

typical kinds

[logical expression]

logical expression

[including inherent limitations]

including inherent limitations

[combination of structures strings]

combination of structures strings

[existing robust parsing models]

existing robust parsing models

[miscommunication reference problems]

miscommunication reference problems

[outputs from information extraction IE]

outputs from information extraction IE

[reflecting the social relationship ranks]

reflecting the social relationship ranks

[speech processing]

speech processing

[grammaticality correct sentences equivalence in meaning]

grammaticality correct sentences equivalence in meaning

[features e .g]

features e .g

[general-purpose NLP components]

general-purpose NLP components

[Applications to the representation]

Applications to the representation

[5K vocabulary]

5K vocabulary

[assumption in form]

assumption in form

[solution to word sense induction]

solution to word sense induction

[Senseval-3 Chinese sample task]

Senseval-3 Chinese sample task

[relation to grammar]

relation to grammar

[language for a need]

language for a need

[evaluation of speech]

evaluation of speech

[case-based MT CBMT]

case-based MT CBMT

[parsing of Korean language]

parsing of Korean language

[employing n-gram models for key prediction]

employing n-gram models for key prediction

[gathering a large number]

gathering a large number

[roughly corresponding]

roughly corresponding

[translation of the words]

translation of the words

[recognition of extra-grammatical input]

recognition of extra-grammatical input

[aspects of a parse tree]

aspects of a parse tree

[parsing-as-deduction in a resource]

parsing-as-deduction in a resource

[kinds of phrases]

kinds of phrases

[a attributes such as size]

a attributes such as size

[aspects of the performance]

aspects of the performance

[computational theory]

computational theory

[iv fine-grained modeling]

iv fine-grained modeling

[effect of features]

effect of features

[tagger use of tag contexts]

tagger use of tag contexts

[small corpus ,000 words]

small corpus ,000 words

[110 ,000 words]

110 ,000 words

[corpus of translations]

corpus of translations

[model of machine translation]

model of machine translation

[string comparison methods]

string comparison methods

[automatic translation of language]

automatic translation of language

[clauses in a program]

clauses in a program

[use in word n-grams]

use in word n-grams

[efficiency gains for the new algorithm]

efficiency gains for the new algorithm

[basics of UCG]

basics of UCG

[interpolation methods like interpolation]

interpolation methods like interpolation

[determining the literal meaning from users]

determining the literal meaning from users

[numerous aspects such as communication procedures]

numerous aspects such as communication procedures

[numeric-valued attributes such as size perspective-taking]

numeric-valued attributes such as size perspective-taking

[system for grammar development]

system for grammar development

[fine-grained modeling of word features]

fine-grained modeling of word features

[according to needs]

according to needs

[word alignment techniques]

word alignment techniques

[generic mathematical formalism]

generic mathematical formalism

[.284 SRHs]

.284 SRHs

[performance for representations]

performance for representations

[standard translation task]

standard translation task

[enumerating all indexings]

enumerating all indexings

[verbal interaction with knowledge]

verbal interaction with knowledge

[Markov models]

Markov models

[collocation observation]

collocation observation

[new problems including unlimited vocabulary]

new problems including unlimited vocabulary

[optimal indexation algorithm]

optimal indexation algorithm

[use of tag contexts]

use of tag contexts

[candidates for understanding results]

candidates for understanding results

[symbolic resources]

symbolic resources

[competing]

competing

[free indexation algorithm]

free indexation algorithm

[extensional equivalence of concepts]

extensional equivalence of concepts

[Using a state-of-the-art Chinese word sense]

Using a state-of-the-art Chinese word sense

[NE .g he/she/man/woman]

NE .g he/she/man/woman

[priors]

priors

[containing ,449 word tokens]

containing ,449 word tokens

[identification of Chinese names]

identification of Chinese names

[new representation]

new representation

[different strategies one utilizing mechanisms]

different strategies one utilizing mechanisms

[Role]

Role

[new systems]

new systems

[final language]

final language

[first-order class grammar]

first-order class grammar

[approach to academic writing]

approach to academic writing

[context features]

context features

[model parameters than models]

model parameters than models

[corpus words]

corpus words

[Educational Research Methods]

Educational Research Methods

[validity of complex databases]

validity of complex databases

[grammar development]

grammar development

[meta-inference]

meta-inference

[spectral mapping]

spectral mapping

[respects]

respects

[straight C code]

straight C code

[applying a set to entries]

applying a set to entries

[analysis for a large class]

analysis for a large class

[post-processing the output]

post-processing the output

[individual theories]

individual theories

[tag contexts]

tag contexts

[more frequently predictors of document descriptors]

more frequently predictors of document descriptors

[algorithm automatically acquiring new stems]

algorithm automatically acquiring new stems

[systematic patterns in translation data]

systematic patterns in translation data

[block bigram model]

block bigram model

[char t]

char t

[improvement over traditional word alignment techniques]

improvement over traditional word alignment techniques

[WSD evaluation methodology]

WSD evaluation methodology

[word corpus]

word corpus

[spatial constraints from the text]

spatial constraints from the text

[areas of natural language processing]

areas of natural language processing

[many NLP problems]

many NLP problems

[separate components]

separate components

[search engine]

search engine

[inductive decision tree learning]

inductive decision tree learning

[Syllogistic reasoning]

Syllogistic reasoning

[99% accuracy in language identification]

99% accuracy in language identification

[attribute sequences]

attribute sequences

[ATIS Data COllection]

ATIS Data COllection

[Lucy a English understanding system]

Lucy a English understanding system

[also a drastic generalization corresponding]

also a drastic generalization corresponding

[efficiency gains for the algorithm]

efficiency gains for the algorithm

[speech/language integration architecture]

speech/language integration architecture

[recognition of input]

recognition of input

[individually good predictors of document descriptors]

individually good predictors of document descriptors

[gathering a large number of abstracts]

gathering a large number of abstracts

[prototype concordancer]

prototype concordancer

[WI systems]

WI systems

[bearing on reasoning]

bearing on reasoning

[protocol analysis]

protocol analysis

[TDMT the most dominant part]

TDMT the most dominant part

[F-measure a decrease in F-measure error]

F-measure a decrease in F-measure error

[such as communication procedures]

such as communication procedures

[temporary belief]

temporary belief

[Dependency Diagram]

Dependency Diagram

[Schutze's idea of evaluation]

Schutze's idea of evaluation

[crucial issue]

crucial issue

[methodology automatically training SPoT]

methodology automatically training SPoT

[making analysis more robust---an scheduling parser]

making analysis more robust---an scheduling parser

[extent entailment]

extent entailment

[several language models LMs]

several language models LMs

[more complex mixtures]

more complex mixtures

[using the grammar formalism]

using the grammar formalism

[social relationship ranks]

social relationship ranks

[framework restrictive than ones]

framework restrictive than ones

[phrase in another language]

phrase in another language

[complexity of unification]

complexity of unification

[MT research]

MT research

[involving automatic extraction from text]

involving automatic extraction from text

[well-formedness conditions]

well-formedness conditions

[problems of vagueness]

problems of vagueness

[determining the meaning]

determining the meaning

[execution]

execution

[coherence rules]

coherence rules

[Chinese word sense disambiguation model]

Chinese word sense disambiguation model

[using likelihood method]

using likelihood method

[speech/language]

speech/language

[superiority over existing methods]

superiority over existing methods

[unannotated text]

unannotated text

[disambiguation WSD models]

disambiguation WSD models

[two problems a numeric-valued attributes]

two problems a numeric-valued attributes

[node in a graph]

node in a graph

[several machine translation tests]

several machine translation tests

[three dimensions skill level]

three dimensions skill level

[word segmenter]

word segmenter

[test suite]

test suite

[string with the best performance]

string with the best performance

[many possible structures]

many possible structures

[training factors on performance]

training factors on performance

[contributions]

contributions

[place for the analysis component]

place for the analysis component

[tagging LMs]

tagging LMs

[foundation exploring bitexts]

foundation exploring bitexts

[3 indicators]

3 indicators

[Automatic Speech Recognition technology]

Automatic Speech Recognition technology

[advances in phrasal translation]

advances in phrasal translation

[unsupervised learning algorithm]

unsupervised learning algorithm

[process in understanding of language]

process in understanding of language

[application of existing metrics]

application of existing metrics

[utilizing primarily knowledge-based mechanisms]

utilizing primarily knowledge-based mechanisms

[instantiation of clauses]

instantiation of clauses

[domain strategy]

domain strategy

[event type networks]

event type networks

[including jointly conditioning on consecutive words]

including jointly conditioning on consecutive words

[topmost]

topmost

[superiority]

superiority

[suitable superset]

suitable superset

[annotators]

annotators

[extracting information from unrestricted texts]

extracting information from unrestricted texts

[dynamic combiner]

dynamic combiner

[evidence from an 500 features]

evidence from an 500 features

[key factors behind this finding]

key factors behind this finding

[parsing choices]

parsing choices

[using rules]

using rules

[common ground in interaction]

common ground in interaction

[approach to Web-based writing]

approach to Web-based writing

[inflection such as English]

inflection such as English

[automated evaluation techniques]

automated evaluation techniques

[CCLINC]

CCLINC

[meteorological events]

meteorological events

[model of statistical machine translation]

model of statistical machine translation

[three dimensions]

three dimensions

[including reference failures]

including reference failures

[developing the grammars in parallel]

developing the grammars in parallel

[research on problems]

research on problems

[levels of performance]

levels of performance

[open-domain question answering capability]

open-domain question answering capability

[utterance classification]

utterance classification

[relatively domain-independent capabilities]

relatively domain-independent capabilities

[text-understanding]

text-understanding

[robust speech dialogue interactions]

robust speech dialogue interactions

[collection]

collection

[small amount of speech]

small amount of speech

[target recognition]

target recognition

[features of questions]

features of questions

[pattern-matching parser]

pattern-matching parser

[instance MT systems]

instance MT systems

[scale of spelling-checkers]

scale of spelling-checkers

[template-based generation component]

template-based generation component

[conglomeration of individual lexical rules]

conglomeration of individual lexical rules

[Complete distribution of grammatical knowledge]

Complete distribution of grammatical knowledge

[considerable importance to Machine Translation]

considerable importance to Machine Translation

[technology of powerful computers]

technology of powerful computers

[possible candidates the structure]

possible candidates the structure

[loglinear models]

loglinear models

[explicit use]

explicit use

[including a word on APs]

including a word on APs

[default reasoning]

default reasoning

[identifying words]

identifying words

[zero occurrences]

zero occurrences

[contributing in the sentence]

contributing in the sentence

[speed of the program]

speed of the program

[novel approach to machine translation]

novel approach to machine translation

[phonological system speech recognition]

phonological system speech recognition

[NE e he/she/man/woman]

NE e he/she/man/woman

[syntactic categories]

syntactic categories

[phrase structure]

phrase structure

[combining several language models]

combining several language models

[applications of some]

applications of some

[generative phonological approach]

generative phonological approach

[neural network]

neural network

[challenge extending the CSR corpus]

challenge extending the CSR corpus

[improvement in alignment quality]

improvement in alignment quality

[consisting]

consisting

[sublanguages with text structure]

sublanguages with text structure

[answers in multiple corpora]

answers in multiple corpora

[resultant corpus]

resultant corpus

[paraphrase classification accuracy]

paraphrase classification accuracy

[novel bootstrapping approach tagging]

novel bootstrapping approach tagging

[sentences in comparable non-parallel corpora]

sentences in comparable non-parallel corpora

[F-measure a decrease]

F-measure a decrease

[statistical machine translation MT]

statistical machine translation MT

[broad range]

broad range

[certain aspects]

certain aspects

[segments of descriptions]

segments of descriptions

[Senseval-3 sample task]

Senseval-3 sample task

[identifying words in Chinese newspapers]

identifying words in Chinese newspapers

[important task]

important task

[character size]

character size

[parsing-as-deduction]

parsing-as-deduction

[equivalence of the two formalisms]

equivalence of the two formalisms

[corresponding to semantic domains]

corresponding to semantic domains

[asynchronous message passing]

asynchronous message passing

[problems a numeric-valued attributes]

problems a numeric-valued attributes

[maximum a estimation of Markov models]

maximum a estimation of Markov models

[accuracy superior]

accuracy superior

[probabilistic grammars]

probabilistic grammars

[method building corpus]

method building corpus

[inducing semantic verb classes]

inducing semantic verb classes

[approach to Web-based academic writing]

approach to Web-based academic writing

[linguistic expressiveness]

linguistic expressiveness

[natural sublanguages]

natural sublanguages

[handling words]

handling words

[learning phrases longer than three words]

learning phrases longer than three words

[approach for Interactive Machine Translation]

approach for Interactive Machine Translation

[minutes per extract]

minutes per extract

[SI training of Markov models]

SI training of Markov models

[using bilingual corpora a resource]

using bilingual corpora a resource

[i parsing a verb language]

i parsing a verb language

[Most importantly texts]

Most importantly texts

[32 improvement]

32 improvement

[Heuristic Parsing Model]

Heuristic Parsing Model

[indicators for the prediction task]

indicators for the prediction task

[probabilistic character recognition model]

probabilistic character recognition model

[system for scoring sets]

system for scoring sets

[concurrent computation model]

concurrent computation model

[also a generalization of chart Parsing]

also a generalization of chart Parsing

[project stage]

project stage

[evaluation using Arabic information retrieval]

evaluation using Arabic information retrieval

[using bilingual corpora a available resource]

using bilingual corpora a available resource

[multilingual input]

multilingual input

[role conveying the meaning]

role conveying the meaning

[aggregate]

aggregate

[several translation models]

several translation models

[state proprietary Arabic stemmer]

state proprietary Arabic stemmer

[holding candidates]

holding candidates

[clustering methods]

clustering methods

[binary relational calculus notation]

binary relational calculus notation

[using a model in place]

using a model in place

[argumentation]

argumentation

[.3M characters]

.3M characters

[system development to new domains]

system development to new domains

[properties of discourse]

properties of discourse

[intelligent mobile agents]

intelligent mobile agents

[document descriptors keywords]

document descriptors keywords

[lexical features from the contexts]

lexical features from the contexts

[using corpora a resource]

using corpora a resource

[using interpolation]

using interpolation

[analysis of people's answers]

analysis of people's answers

[new problems including vocabulary]

new problems including vocabulary

[exploiting a non-parallel corpus]

exploiting a non-parallel corpus

[concepts of the theory]

concepts of the theory

[settings]

settings

[role in human cognition]

role in human cognition

[new approach for Machine Translation]

new approach for Machine Translation

[drastic generalization roughly corresponding]

drastic generalization roughly corresponding

[segment order]

segment order

[representation the Phrasal Lexicon DHPL]

representation the Phrasal Lexicon DHPL

[predicting segment boundaries in multiparty dialogue]

predicting segment boundaries in multiparty dialogue

[scheme's ability]

scheme's ability

[large non-parallel corpus]

large non-parallel corpus

[using bilingual corpora]

using bilingual corpora

[well-formedness conditions on trees]

well-formedness conditions on trees

[features over parse trees]

features over parse trees

[improvement in average precision]

improvement in average precision

[training SPoT]

training SPoT

[additional increase]

additional increase

[building corpus]

building corpus

[placement]

placement

[improvements in the BLEU scores]

improvements in the BLEU scores

[89 .75% F-measure a relative decrease]

89 .75% F-measure a relative decrease

[Phrasal Lexicon DHPL]

Phrasal Lexicon DHPL

[constraints in broad domains]

constraints in broad domains

[rightward manner]

rightward manner

[use including conditioning]

use including conditioning

[describing the processing of utterances]

describing the processing of utterances

[three separate components]

three separate components

[Robust parsing a verb final language]

Robust parsing a verb final language

[Question Answering QA systems]

Question Answering QA systems

[usual pooling]

usual pooling

[need developing time approximations]

need developing time approximations

[IDUS development]

IDUS development

[development model constructing a system]

development model constructing a system

[new methodology]

new methodology

[effect using an bilingual corpus]

effect using an bilingual corpus

[single frame]

single frame

[set of extraction]

set of extraction

[two domains]

two domains

[valid dictionary]

valid dictionary

[present language search engines]

present language search engines

[Dynamic Hierarchical Phrasal Lexicon [Zernik88]]

Dynamic Hierarchical Phrasal Lexicon [Zernik88]

[technique learning morphology]

technique learning morphology

[various relations]

various relations

[problem enumerating all indexings]

problem enumerating all indexings

[recent improvements]

recent improvements

[Acyclic Graph Kernel]

Acyclic Graph Kernel

[knowledge-based mechanisms]

knowledge-based mechanisms

[fair copy]

fair copy

[information retrieval techniques]

information retrieval techniques

[ambiguity packing techniques for Grammars]

ambiguity packing techniques for Grammars

[methods .e]

methods .e

[interactions of task-oriented dialogs]

interactions of task-oriented dialogs

[role in the determination]

role in the determination

[Using utterances]

Using utterances

[use in case-based MT CBMT]

use in case-based MT CBMT

[frame encoding the full meaning]

frame encoding the full meaning

[conglomeration]

conglomeration

[most approaches]

most approaches

[paraphrases in language]

paraphrases in language

[three techniques making syntactic analysis]

three techniques making syntactic analysis

[interaction with knowledge]

interaction with knowledge

[standard outputs systems entity annotations]

standard outputs systems entity annotations

[using additional features of the tree]

using additional features of the tree

[large non-parallel newspaper corpora]

large non-parallel newspaper corpora

[theory of discourse structure]

theory of discourse structure

[Semantic Role System]

Semantic Role System

[real dialogue data]

real dialogue data

[additional 500 ,000 features]

additional 500 ,000 features

[use of features including conditioning]

use of features including conditioning

[evaluating the effectiveness]

evaluating the effectiveness

[WSD programs]

WSD programs

[performing inference]

performing inference

[IDUS]

IDUS

[common noun]

common noun

[mixture language model]

mixture language model

[words from high-accuracy word-level alignment models]

words from high-accuracy word-level alignment models

[making analysis]

making analysis

[monitoring the collection of utterances]

monitoring the collection of utterances

[new users]

new users

[entity-oriented language definition]

entity-oriented language definition

[three language system domains]

three language system domains

[addressee identification meetings]

addressee identification meetings

[multiparty dialogue]

multiparty dialogue

[advantage of linguistic information]

advantage of linguistic information

[Robust efficient parsing of Korean]

Robust efficient parsing of Korean

[bilingual material the Hansards]

bilingual material the Hansards

[candidate parses]

candidate parses

[producing phrases from a corpora]

producing phrases from a corpora

[purpose in discourse]

purpose in discourse

[transfer mappings]

transfer mappings

[using semantic categories on the terminal]

using semantic categories on the terminal

[approach for Machine Translation]

approach for Machine Translation

[novel method discovering parallel sentences]

novel method discovering parallel sentences

[areas of language processing]

areas of language processing

[Rapid system development via knowledge-based acquisition]

Rapid system development via knowledge-based acquisition

[alternative speech recognition hypotheses]

alternative speech recognition hypotheses

[machine-readable dictionaries MRD's]

machine-readable dictionaries MRD's

[segment methods]

segment methods

[defeat]

defeat

[parse-based evaluation]

parse-based evaluation

[examination of the complexity]

examination of the complexity

[speaking]

speaking

[different language pairs]

different language pairs

[advances in Speech Recognition technology]

advances in Speech Recognition technology

[CFG filtering techniques]

CFG filtering techniques

[combination of hand-built resources]

combination of hand-built resources

[two contributions speech recognition]

two contributions speech recognition

[many theories]

many theories

[Wall Street]

Wall Street

[89 .75% F-measure]

89 .75% F-measure

[dynamic word]

dynamic word

[using corpora a more available resource]

using corpora a more available resource

[rating]

rating

[current state]

current state

[speech recognition of a natural language]

speech recognition of a natural language

[including limitations of MT architectures]

including limitations of MT architectures

[three methods .e methods]

three methods .e methods

[dialogue duration]

dialogue duration

[graphs indispensable to papers]

graphs indispensable to papers

[simple MT-based paraphrasing technique]

simple MT-based paraphrasing technique

[performance enhancing]

performance enhancing

[reflecting the social relationship e .g]

reflecting the social relationship e .g

[roles in the frame semantics paradigm]

roles in the frame semantics paradigm

[evidence from the input]

evidence from the input

[four speech recognition applications]

four speech recognition applications

[amount of theoretical work]

amount of theoretical work

[conventional word-trigram recognition requiring transcription]

conventional word-trigram recognition requiring transcription

[system combination]

system combination

[maintaining the scheme's ability]

maintaining the scheme's ability

[Using alignment techniques]

Using alignment techniques

[expressions of languages to semantics]

expressions of languages to semantics

[less precise adverbs]

less precise adverbs

[important issue]

important issue

[making syntactic analysis]

making syntactic analysis

[scheduling parser]

scheduling parser

[Unification Grammar UCG]

Unification Grammar UCG

[grammaticality at least 99% sentences equivalence]

grammaticality at least 99% sentences equivalence

[syntactic analysis robust---an agenda-based scheduling parser]

syntactic analysis robust---an agenda-based scheduling parser

[Bayes]

Bayes

[domain models with off-the-shelf classifiers]

domain models with off-the-shelf classifiers

[utilizing knowledge-based mechanisms]

utilizing knowledge-based mechanisms

[balance matching]

balance matching

[principled techniques]

principled techniques

[integrating automatic Q/A applications into environments]

integrating automatic Q/A applications into environments

[combining knowledge]

combining knowledge

[training factors on predictive performance]

training factors on predictive performance

[beginning of the sentence]

beginning of the sentence

[identification of names]

identification of names

[complex mixtures of techniques]

complex mixtures of techniques

[NE e]

NE e

[problem of the detection]

problem of the detection

[using a relational calculus notation]

using a relational calculus notation

[marine weather]

marine weather

[constraints as levels of representation]

constraints as levels of representation

[correct sentences]

correct sentences

[enrichment]

enrichment

[integrating Q/A applications into environments]

integrating Q/A applications into environments

[recent advances]

recent advances

[computer-assisted academic writing]

computer-assisted academic writing

[Complete lexical distribution of knowledge]

Complete lexical distribution of knowledge

[performance (typically]

performance (typically

[understanding technology]

understanding technology

[polynomial time approximations]

polynomial time approximations

[construction as basis]

construction as basis

[graph's]

graph's

[three methods constructing word vectors .e]

three methods constructing word vectors .e

[Arabic's]

Arabic's

[grammar checker]

grammar checker

[frequently good predictors]

frequently good predictors

[simpler set]

simpler set

[novel IE paradigm]

novel IE paradigm

[types of the errors]

types of the errors

[state stemmer]

state stemmer

[predicting the translation of the words]

predicting the translation of the words

[way identifying argument structures]

way identifying argument structures

[efficacy applying evaluation techniques]

efficacy applying evaluation techniques

[interactive question-answering Q/A system]

interactive question-answering Q/A system

[generation component]

generation component

[126 ,610 sentences]

126 ,610 sentences

[evidence from an features]

evidence from an features

[contributing to the word matches]

contributing to the word matches

[ATN]

ATN

[system output]

system output

[portion using a symbol]

portion using a symbol

[correctness]

correctness

[space of inference types]

space of inference types

[neat form]

neat form

[language modeling]

language modeling

[evaluating QA system performance]

evaluating QA system performance

[different strategies one utilizing knowledge-based mechanisms]

different strategies one utilizing knowledge-based mechanisms

[bilingual corpora a commonly available resource]

bilingual corpora a commonly available resource

[generalization of chart Parsing corresponding]

generalization of chart Parsing corresponding

[overlapping]

overlapping

[i efficient parsing language]

i efficient parsing language

[single tagging result]

single tagging result

[.0% relative improvement over system]

.0% relative improvement over system

[.5% agreement with a state]

.5% agreement with a state

[predicting segment boundaries]

predicting segment boundaries

[for scoring sets on the basis]

for scoring sets on the basis

[parallel sentences]

parallel sentences

[increasing parsing time complexity]

increasing parsing time complexity

[assignment of descriptors]

assignment of descriptors

[revision support systems]

revision support systems

[iv modeling]

iv modeling

[using a POS tagger]

using a POS tagger

[two new disambiguation systems one]

two new disambiguation systems one

[general-purpose language understanding technology]

general-purpose language understanding technology

[re-utterance]

re-utterance

[underpinning]

underpinning

[knowledge-based acquisition]

knowledge-based acquisition

[Word]

Word

[usage]

usage

[combination of hand-built symbolic resources]

combination of hand-built symbolic resources

[additional source]

additional source

[several kinds of knowledge]

several kinds of knowledge

[two distinct domains]

two distinct domains

[using triplets of words]

using triplets of words

[adopting a MT-based paraphrasing technique]

adopting a MT-based paraphrasing technique

[bitexts of the laws]

bitexts of the laws

[combination of symbolic resources]

combination of symbolic resources

[posteriori estimation]

posteriori estimation

[optimal free indexation algorithm]

optimal free indexation algorithm

[Grammars LFG]

Grammars LFG

[generalization corresponding]

generalization corresponding

[significant efficiency gains for the algorithm]

significant efficiency gains for the algorithm

[state-of-the-art evaluation measures for machine translation]

state-of-the-art evaluation measures for machine translation

[insight into the potential]

insight into the potential

[starting with a small parallel corpus]

starting with a small parallel corpus

[containing word tokens]

containing word tokens

[progressing from generation of true text]

progressing from generation of true text

[CARE]

CARE

[easy exploitation]

easy exploitation

[building upon the head-oriented notions]

building upon the head-oriented notions

[utterances of speech from sites]

utterances of speech from sites

[sentence-plan-generator SPG]

sentence-plan-generator SPG

[bilingual corpora a resource]

bilingual corpora a resource

[corpora a more commonly available resource]

corpora a more commonly available resource

[target language word]

target language word

[robust parsing models]

robust parsing models

[words contributing]

words contributing

[techniques making analysis robust---an scheduling parser]

techniques making analysis robust---an scheduling parser

[contextual clues such as editing terms]

contextual clues such as editing terms

[special type]

special type

[larger improvements]

larger improvements

[two systems one]

two systems one

[linguistic tools]

linguistic tools

[Robust efficient parsing language]

Robust efficient parsing language

[sentence alignment]

sentence alignment

[involving extraction]

involving extraction

[machine translation tool kit]

machine translation tool kit

[two successive learners]

two successive learners

[samples from an entity-oriented language definition]

samples from an entity-oriented language definition

[acquiring new stems]

acquiring new stems

[seed sentences]

seed sentences

[disambigation]

disambigation

[ways dealing]

ways dealing

[specific move]

specific move

[understanding of language]

understanding of language

[including conditioning on words]

including conditioning on words

[speculative claims to the contrary]

speculative claims to the contrary

[better indicators for the prediction task]

better indicators for the prediction task

[Hierarchical Graph HDAG Kernel]

Hierarchical Graph HDAG Kernel

[template-based approaches]

template-based approaches

[IBM MT system]

IBM MT system

[combinatorics of free indexation]

combinatorics of free indexation

[attribute sequences of the HDAGs]

attribute sequences of the HDAGs

[German corpus of SRHs]

German corpus of SRHs

[leveraging language technology]

leveraging language technology

[transition graphs]

transition graphs

[processing of utterances]

processing of utterances

[especially on the changes]

especially on the changes

[predicate with honorifics]

predicate with honorifics

[about 110 words]

about 110 words

[exploring English-Chinese bitexts in a volume]

exploring English-Chinese bitexts in a volume

[according to the precision metric]

according to the precision metric

[using parsing strategies]

using parsing strategies

[complex databases]

complex databases

[features for paraphrase classification]

features for paraphrase classification

[using a little speech from speakers]

using a little speech from speakers

[acquisition of grammars]

acquisition of grammars

[new way identifying argument structures]

new way identifying argument structures

[size of the databases]

size of the databases

[aiming for complete interpretations]

aiming for complete interpretations

[word sense WSD]

word sense WSD

[35 .0% relative improvement]

35 .0% relative improvement

[examination of the computational complexity]

examination of the computational complexity

[expressions of natural languages]

expressions of natural languages

[explicit information from dictionaries MRD's]

explicit information from dictionaries MRD's

[output of tools]

output of tools

[meaning equivalence]

meaning equivalence

[drafts]

drafts

[equivalence of the formalisms]

equivalence of the formalisms

[problem predicting segment boundaries]

problem predicting segment boundaries

[harvesting English-Chinese bitexts of the laws]

harvesting English-Chinese bitexts of the laws

[RCG]

RCGs, RCG

[unsegmented Arabic corpus]

unsegmented Arabic corpus

[building a stemmer]

building a stemmer

[individually high-precision rules]

individually high-precision rules

[best performance (typically]

best performance (typically

[evaluation of word sense algorithms]

evaluation of word sense algorithms

[fundamentally different strategies utilizing mechanisms]

fundamentally different strategies utilizing mechanisms

[first-order statistical class grammar]

first-order statistical class grammar

[omissions]

omissions

[parse reduction]

parse reduction

[ACE]

ACE

[amount of variation superior]

amount of variation superior

[Results from experiments]

Results from experiments

[translation memory]

translation memory

[constructing a phonological system speech recognition]

constructing a phonological system speech recognition

[attributes such as size]

attributes such as size

[semantic representation USR]

semantic representation USR

[most men like women]

most men like women

[visualization]

visualization

[precise adverbs]

precise adverbs

[parsing of Korean final language]

parsing of Korean final language

[usually in a manner]

usually in a manner

[error reduction of 4]

error reduction of 4

[two standard outputs]

two standard outputs

[analysis component]

analysis component

[output of linguistic tools]

output of linguistic tools

[many questions]

many questions

[ambiguity packing disambiguation techniques]

ambiguity packing disambiguation techniques

[training resources]

training resources

[determining the meaning from users]

determining the meaning from users

[i Robust efficient parsing language]

i Robust efficient parsing language

[parsing choices of the parser]

parsing choices of the parser

[methods for scoring system output]

methods for scoring system output

[best strategies]

best strategies

[parallel corpus ,000 words]

parallel corpus ,000 words

[experiments corresponding to domains]

experiments corresponding to domains

[translation of natural language]

translation of natural language

[continuous Markov models]

continuous Markov models

[providing insights]

providing insights

[confines]

confines

[three methods]

three methods

[parser/generator for LFG]

parser/generator for LFG

[representing the meaning through the use]

representing the meaning through the use

[conversion]

conversion

[fine-grained modeling of unknown word features]

fine-grained modeling of unknown word features

[abstracts for learning]

abstracts for learning

[language search engines]

language search engines

[new technology of personal computers]

new technology of personal computers

[measure with judgment]

measure with judgment

[research on new problems including vocabulary]

research on new problems including vocabulary

[building the editor]

building the editor

[extensions of pre-existing components]

extensions of pre-existing components

[use including conditioning on words]

use including conditioning on words

[a attributes]

a attributes

[statistics of models]

statistics of models

[model using associative processors]

model using associative processors

[samples from an language definition]

samples from an language definition

[speed-up]

speed-up

[SemCor data sets]

SemCor data sets

[unique properties of grammars TAG]

unique properties of grammars TAG

[passing protocols]

passing protocols

[automatically predicting segment boundaries]

automatically predicting segment boundaries

[predicate argument structures]

predicate argument structures

[application back]

application back

[surface English]

surface English

[variations in the phrasing]

variations in the phrasing

[domain acoustic models with classifiers]

domain acoustic models with classifiers

[theories of anaphora resolution]

theories of anaphora resolution

[complete sentences]

complete sentences

[numbering system]

numbering system

[neighbour]

neighbour

[grammaticality 99% sentences equivalence in meaning]

grammaticality 99% sentences equivalence in meaning

[corpora a much more available resource]

corpora a much more available resource

[features of the tree]

features of the tree

[transition graphs for a type]

transition graphs for a type

[two contributions continuous speech recognition]

two contributions continuous speech recognition

[changes]

changes

[many other NLP problems]

many other NLP problems

[primarily on explicit information from dictionaries]

primarily on explicit information from dictionaries

[Markov probability]

Markov probability

[expressiveness of the formalisms]

expressiveness of the formalisms

[disambiguation systems]

disambiguation systems

[starting with a small corpus]

starting with a small corpus

[.0% relative improvement]

.0% relative improvement

[primarily on information]

primarily on information

[one element]

one element

[identifying unknown words]

identifying unknown words

[building a Arabic stemmer]

building a Arabic stemmer

[languages with inflection]

languages with inflection

[word forms]

word forms

[large Chinese newspaper corpora]

large Chinese newspaper corpora

[connecting referents]

connecting referents

[machine dictionaries]

machine dictionaries

[interactive method]

interactive method

[analysis of the English coordinate sentences]

analysis of the English coordinate sentences

[definition of Minimalist grammars]

definition of Minimalist grammars

[knowledge about knowledge]

knowledge about knowledge

[learning phrases words]

learning phrases words

[using Network classifiers]

using Network classifiers

[two word-sense systems]

two word-sense systems

[information from dictionaries]

information from dictionaries

[building spelling-checkers for languages]

building spelling-checkers for languages

[software infrastructure]

software infrastructure

[omissions of arguments]

omissions of arguments

[measure with human judgment]

measure with human judgment

[architecture with a variety]

architecture with a variety

[iii system development to domains]

iii system development to domains

[problem combining language models LMs]

problem combining language models LMs

[natural sublanguages with text structure]

natural sublanguages with text structure

[edges with a valid semantic interpretation]

edges with a valid semantic interpretation

[SRHs]

SRHs

[browsing the repository]

browsing the repository

[independent event-based extractive summarization approaches]

independent event-based extractive summarization approaches

[aspects of language learning]

aspects of language learning

[answering capability on text]

answering capability on text

[problems numeric-valued attributes such as size]

problems numeric-valued attributes such as size

[attributes such as size b perspective-taking]

attributes such as size b perspective-taking

[fully-connected statistical class grammar]

fully-connected statistical class grammar

[Hierarchical Acyclic Graph HDAG]

Hierarchical Acyclic Graph HDAG

[small parallel corpus words]

small parallel corpus words

[probabilistic character recognition OCR model]

probabilistic character recognition OCR model

[use including conditioning on consecutive words]

use including conditioning on consecutive words

[i efficient parsing of Korean language]

i efficient parsing of Korean language

[bigram language]

bigram language

[practical ways]

practical ways

[grammatical formalisms Tree Grammars]

grammatical formalisms Tree Grammars

[progressing into the noisy output]

progressing into the noisy output

[experience with TACITUS]

experience with TACITUS

[lack of methods]

lack of methods

[provably optimal indexation algorithm]

provably optimal indexation algorithm

[class grammar]

class grammar

[using an corpus]

using an corpus

[several key factors including limitations]

several key factors including limitations

[powerful tools]

powerful tools

[mission]

mission

[simple means heuristic learning]

simple means heuristic learning

[unsupervised algorithm acquiring stems]

unsupervised algorithm acquiring stems

[effects of segment order]

effects of segment order

[reappraising these counterexamples]

reappraising these counterexamples

[unique author]

unique author

[several key factors including inherent limitations]

several key factors including inherent limitations

[data sources]

data sources

[basic goal]

basic goal

[paradigm for interaction]

paradigm for interaction

[modeling of unknown word features]

modeling of unknown word features

[sum]

sum

[system speech recognition]

system speech recognition

[sentences in non-parallel corpora]

sentences in non-parallel corpora

[part of text processing software]

part of text processing software

[output of an existing parser]

output of an existing parser

[part of a study]

part of a study

[therefore a good choice]

therefore a good choice

[constraints from the text]

constraints from the text

[application of Unification Categorial Grammar]

application of Unification Categorial Grammar

[i efficient parsing]

i efficient parsing

[large amount of work]

large amount of work

[interactions of dialogs]

interactions of dialogs

[intention component]

intention component

[real-time demonstration]

real-time demonstration

[sentence-plan-generator]

sentence-plan-generator

[using features as evidence]

using features as evidence

[analysis more robust---an scheduling parser]

analysis more robust---an scheduling parser

[standard use of BLEU]

standard use of BLEU

[TV shows]

TV shows

[story]

story

[symbol]

symbol

[bus]

bus

[achieving similar performance to mixtures]

achieving similar performance to mixtures

[frequency]

frequency

[ii quality translation]

ii quality translation

[missiles]

missiles

[SCF data]

SCF data

[generating]

generating

[parallel corpora a resource]

parallel corpora a resource

[finding more data]

finding more data

[training successive learners]

training successive learners

[practical efficiency]

practical efficiency

[applying models]

applying models

[planning COMPLEX a computational lexicon]

planning COMPLEX a computational lexicon

[detailed linguistic information]

detailed linguistic information

[CCLINC Coalition Language System]

CCLINC Coalition Language System

[computer-assisted writing]

computer-assisted writing

[extensions of components]

extensions of components

[newswire text]

newswire text

[using seeds]

using seeds

[issue in Chinese Language Processing]

issue in Chinese Language Processing

[empirical MT research]

empirical MT research

[overt case markers]

overt case markers

[Prolog cf query optimisation]

Prolog cf query optimisation

[framework less restrictive]

framework less restrictive

[ambiguity resolution]

ambiguity resolution

[using strategies]

using strategies

[Travel]

Travel

[posing problems for readers]

posing problems for readers

[strength of antecedence]

strength of antecedence

[effectiveness of this approach]

effectiveness of this approach

[using less memory]

using less memory

[triplets]

triplets

[starting with a parallel corpus words]

starting with a parallel corpus words

[Two styles performing inference]

Two styles performing inference

[Lincoln Laboratory]

Lincoln Laboratory

[men like mostly young women]

men like mostly young women

[speaker group]

speaker group

[Robust parsing of Korean final language]

Robust parsing of Korean final language

[Grammars approach]

Grammars approach

[harvesting English-Chinese bitexts from the Web]

harvesting English-Chinese bitexts from the Web

[attaining a design overheads]

attaining a design overheads

[responses to each user]

responses to each user

[interesting combination of symbolic resources]

interesting combination of symbolic resources

[example from an bi-directional fragment]

example from an bi-directional fragment

[a numeric-valued attributes]

a numeric-valued attributes

[language question]

language question

[good predictors of document descriptors]

good predictors of document descriptors

[generative probabilistic model of language]

generative probabilistic model of language

[discourse phenomena]

discourse phenomena

[grammar generation]

grammar generation

[block selection]

block selection

[different ways]

different ways

[computational theory of reasoning]

computational theory of reasoning

[iii Rapid system development via acquisition]

iii Rapid system development via acquisition

[constructing word vectors]

constructing word vectors

[results of a user evaluation]

results of a user evaluation

[new technology of powerful computers]

new technology of powerful computers

[computers with integral graphics displays]

computers with integral graphics displays

[strategies one utilizing primarily knowledge-based mechanisms]

strategies one utilizing primarily knowledge-based mechanisms

[i parsing of Korean final language]

i parsing of Korean final language

[equivalence in meaning]

equivalence in meaning

[set of extracts]

set of extracts

[sets(]

sets(

[combination strategies]

combination strategies

[behaviour]

behaviour

[including the influence]

including the influence

[modeling of word features]

modeling of word features

[segment order-sensitive methods]

segment order-sensitive methods

[statistics on co-occurrence patterns]

statistics on co-occurrence patterns

[underspecified representation USR]

underspecified representation USR

[preference of approach]

preference of approach

[dependency network]

dependency network

[young men like mostly women]

young men like mostly women

[material Roget's Thesaurus]

material Roget's Thesaurus

[cost of generation systems]

cost of generation systems

[information from machine-readable dictionaries]

information from machine-readable dictionaries

[pooling of the speech data]

pooling of the speech data

[occurrences of a morpheme]

occurrences of a morpheme

[unique properties of grammars]

unique properties of grammars

[priors in models]

priors in models

[relationship between the formalisms]

relationship between the formalisms

[language interface]

language interface, language interfaces

[single semantic frame]

single semantic frame

[computer text]

computer text

[word-level alignment models]

word-level alignment models

[information related]

information related

[representative Chinese-to-English SMT model]

representative Chinese-to-English SMT model

[such MT CBMT]

such MT CBMT

[unique properties]

unique properties

[monitoring the collection of 12,000 utterances]

monitoring the collection of 12,000 utterances

[reliable basis]

reliable basis

[event-based extractive mutli-document summarization approaches]

event-based extractive mutli-document summarization approaches

[integral graphics displays]

integral graphics displays

[alternative speech recognition hypotheses SRH]

alternative speech recognition hypotheses SRH

[including jointly conditioning on multiple words]

including jointly conditioning on multiple words

[four effects the reduction]

four effects the reduction

[one superiority of this method]

one superiority of this method

[Arabic-English translation task]

Arabic-English translation task

[reducing the amount]

reducing the amount

[efficacy]

efficacy

[discourse segmentation]

discourse segmentation

[extensions of this basic idea]

extensions of this basic idea

[directly on word sense]

directly on word sense

[providing insights into the the complexity]

providing insights into the the complexity

