Bulgarian Academy of Sciences

Multi Word Expressions (MWEs) are a key issue and a current weakness for NLP tasks that need some degree of semantic interpretation, with potential contributions to natural language parsing and generation, as well as applications such as machine translation, information retrieval and information extraction. Therefore, a tutorial on this topic is of great relevance for researchers in NLP and related areas.

Attendees to this tutorial should gain insight into linguistic and distributional characteristics of MWEs and, most importantly, their relevance for robust automated natural language processing and language technology, with a thorough overview of methods and resources that support their use. We will provide a theoretical and practical introduction to the topic, with demonstrations of resources and tools available, aiming to equip the attendees with some starting recipes for MWE processing, including tools for the identification and resource construction (e.g. NSP, UCS, mwetoolkit) and annotation (e.g. jMWE). Our target audience includes researchers and practitioners in Language Technology (not necessarily experts in MWEs) who are interested in tasks that involve or could benefit from considering MWEs as a pervasive phenomenon in human language and communication.