The Linguistic Engineering Group

The Linguistic Engineering (LE) Group is part of the Department of Artificial Intelligence at the Institute of Computer Science, Polish Academy of Sciences (IPI PAN).

People

Core team

Associates

Research

The main research areas of the Group

• (Polish) corpus linguistics (National Corpus of Polish), ; cf. the IPI PAN Corpus of Polish and the National Corpus of Polish,

• morphosyntactic tagging and lemmatisation of Polish,
• syntactic and semantic parsing of Polish,
• extraction of linguistic knowledge from corpora,
• information extraction,
• distributional semantics and compositional distributional semantics,
• sentiment analysis,
• credibility assessment of online content,

  * morphosyntactic system of Polish,

• generative linguistic formalisms, esp., HPSG and LFG.

The Group is a member of CLARIN, DARIAH-PL, ELRC, FLaReNet and META-NET.

Current externally funded projects

• CLARIN-PL (Polish chapter of Common Language Resources and Technology Infrastructure)

• CORMETAN (Cognitive and sociocultural analysis of metaphoric expressions in Polish texts)

• CURLICAT (Curated Multilingual Language Resources for CEF AT)

• DARIAH-PL (Digital Research Infrastructure for the Arts and Humanities)

• ELE (European Language Equality)

• ELG (European Language Grid)

• ELRC (European Language Resource Coordination)

• KORBA 2 (Extension of the "Electronic corpus of 17th and 18th century Polish texts" and its integration with the "Electronic Dictionary of the 17th–18th Century Polish")

• HOMADOS (Hampering Misinformation by Assessing Credibility of Online Sources)

• MARCELL (Multilingual Resources for CEF.AT in the legal domain)

• Nexus Linguarum (European network for Web-centred linguistic data science)

• Kwantyfikatory w języku: użycie i znaczenie (Quantifiers in Language: Use and Meaning)

• Scwad (Compositional distributional modelling of Polish language semantics)

• SYNAMET (Microcorpus of Synaesthetic Metaphors. Towards a Formal Description and Efficient Methods of Analysis of Metaphors in Discourse)

Some of our past projects

• CoDeS (Compositional distributional semantic models for identification, discrimination and disambiguation of senses in Polish texts)

• ATLAS (Applied Technology for Language-Aided CMS)

• Automatic detection and correction of annotation errors in Polish language corpora

• Automatic detection of semantic dependencies within verb argument structures in large treebanks

• Automatic extraction of linguistic knowledge from a large corpus of Polish

• CESAR (CEntral and South-east europeAn Resources)

• Chronofleks (A diachronic formal model of Polish inflection and its implementation)

• CLARIN (Polish chapter of Common Language Resources and Technology Infrastructure, see also CLARIN-PL 2)

• Construction of a treebank for Polish using automatic syntactic analysis

• CORE (Computer-based methods for coreference resolution in Polish texts)

• COTHEC (Unified theory of coreference in Polish and its corpus-based verification)

• HPSG Grammar of Polish

• Information Extraction from Polish free text

• The IPI PAN Corpus of Polish

• KORBA (Electronic corpus of 17th and 18th century Polish texts)

• LT4eL (Language Technology for eLearning)

• LUNA (spoken Language UNderstanding in multilinguAl communication systems) with the Polish support

• NEKST (An adaptive system to support problem-solving on the basis of document collections in the Internet)

• NKJP (National Corpus of Polish)

• OPTA (Automatyczne metody rozpoznawania przedmiotów i wyrażeń opinii w języku polskim)

• PARSEME (PARSing and Multi-word Expressions. Towards linguistic precision and computational efficiency in natural language processing)

• Parthenos (Pooling Activities, Resources and Tools for Heritage, E-research Networking, Optimization and Synergies)

• Readability (Measuring the degree of readability of nonliterary Polish texts)

• SYNAT (Creation of a universal, open repository platform for hosting and communication of networked resources of knowledge for science, education and open knowledge-based society)

• Treebank / Test Suite of Polish Utterances

• TextLink (Structuring Discourse in Multilingual Europe)

• TrendMiner (Large-scale, Cross-lingual Trend Mining and Summarisation of Real-time Media Streams)

Publicly available tools and resources

Here are some of the tools and resources created within our projects. See CLIP pages for a more exhaustive list of Polish tools and resources, including more tools and resources developed at ZIL IPI PAN.

Some tools (all open source, under GPL; see also CLIP):

• Morfeusz 2 – a morphological analyser of Polish,

• Spejd – a shallow parsing and disambiguation system,

• Świgra – a DCG parser,

• COMBO – a language-independent system for natural language preprocessing (i.e. morphosyntactic tagging, lemmatisation, dependency parsing and thematic role labelling,

• Concraft — a CRF morphosyntactic tagger of Polish compatible with Morfeusz SGJP,

• PANTERA – a morphosyntactic tagger for Polish,

• TaKIPI – a morphosyntactic tagger for Polish,

• Poliqarp – a corpus indexing and search engine,

• Poliqarp2 – a new generation corpus indexing and search engine,

• Dendrarium – a treebank development system (under development),

• Anotatornia 2 – an annotation tool geared towards historical corpora,

• WSDDE – a system for designing and performing Word Sense Disambiguation experiments,

• Multiservice – web service for various of our tools,

• TermoPL - multiword terms extraction from text

• DSmodels - web service for calculating word similarity using Polish word embeddings

Main resources (many more at CLIP):

• Walenty – a valence dictionary of Polish (described here),

• National Corpus of Polish,

• Polish word embeddings based on NKJP and Wikipedia,

• Polish dependency banks: PDB, PDB-UD, PUD-PL, NKJP1M-UD,

• Dependency parsing models for Polish.

Other activities

Links to some other activities of the Group:

• Journal of Language Modelling

• NLP Seminar at IPI PAN

• PolEval, the evaluation campaign for natural language processing tools for Polish

• conferences organised by the Group:

  • Intelligent Information Systems series of conferences

  • 9th International Conference on Natural Language Processing (PolTAL 2014), 17–19 September 2014, Warsaw, Poland

  • 14th International Workshop on Treebanks and Linguistic Theories (TLT14), 11–12 December 2015, Warsaw, Poland

  • Coreference Resolution Beyond OntoNotes (CORBON 2016) workshop at NAACL 2016 (The 15th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies), 16 June 2016, San Diego, US

  • Joint 2016 Conference on Head-driven Phrase Structure Grammar and Lexical Functional Grammar (HeadLex16), 24–29 July 2016, Warsaw, Poland

  • 2nd Workshop on Coreference Resolution Beyond OntoNotes (CORBON 2017) at EACL 2017 (The 15th Conference of the European Chapter of the Association for Computational Linguistics), 4 April 2017, Valencia, Spain

  • Computational Models of Reference, Anaphora, and Coreference workshop (CRAC) at NAACL 2018 (The 16th Annual Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies), 6 June 2018, New Orleans, USA

  • AI & NLP Workshop Day, 19 October 2018, Warsaw

  • Second Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC 2019), 6 ot 7 June 2019, Minneapolis

  • The 24th International Lexical-Functional Grammar Conference (LFG 2019), 8–10 July 2019, Canberra