PARSEME project

Project factsheet

Polish partners involved

• Institute of Computer Science, Polish Academy of Sciences (Vice Chair of the Action: Adam Przepiórkowski)

Project description

The Action, PARSEME, aims at increasing and enhancing the support of the Europeanmultilingual heritage from Information and Communication Technologies (ICT). Thisgeneral aim is addressed through improving linguistic representativeness, precisionand computational efficiency of Natural Language Processing (NLP) applications. The Action focuses on the major bottleneck of these applications: Multi-Word Expressions(MWEs), i.e. sequences of words with unpredictable properties such as to countsomebody in or to take a haircut. A breakthrough in their modelling and processing canonly result from a coordinated effort of multidisciplinary experts in different languages.COST is the most adequate framework answering this need. Fourteen European languageswill be addressed from a cross-theoretical and cross-methodological perspective, necessaryfor coping with current fragmentation issues. Expected deliverables include enhancedlanguage resources and tools, as well as recommendations of best practices forcutting-edge MWE-aware language models. The Action will lead to a better understandingof the nature of MWEs. It will establish a long-lasting collaboration within amultilingual network of MWE specialists. It will pave the way towards competitive nextgeneration text processing tools which will pay greater attention to languagephenomena.