PARSEME Survey on MWE Resources

Gyri Smørdal Losnegaard (1) , Federico Sangati (2) , Carla Parra Escartín (3) , Agata Savary (4) , Sascha Bargmann (5) , Johanna Monti (6)

1 UiB - University of Bergen
2 FBK - Fondazione Bruno Kessler [Trento, Italy]
3 Hermes Traducciones y Servicios Lingüísticos
4 BDTLN - Bases de données et traitement des langues naturelles
5 Goethe University Frankfurt
6 UNISS - Università degli Studi di Sassari = University of Sassari [Sassari]

This paper summarizes the first results of an ongoing survey on multiword resources carried out within the IC1207 Cost Action PARSEME (PARSing and Multi-word Expressions). Despite the availability of language resource catalogues and the inventory of multiword data-sets available at the SIGLEX-MWE website, multiword resources are scattered and prove to be difficult to be found. In many cases, language resources such as corpora, treebanks or lexical databases include multiwords as part of their data or take them into consideration in their annotations. However, it is needed to centralize these resources so that other researches may subsequently use them. The final aim of this survey is thus to create a portal where researchers may find multiword resources or multiword-aware language resources for their research. We report on how the survey was designed and analyze the data gathered so far. We also discuss the problems we have detected upon examination of the data and possible ways of enhancing the survey.

Mots clés: multiword expressions; language resources; language resource infrastructures

Domaines: Informatique et langage [cs.CL]

Gyri Smørdal Losnegaard, Federico Sangati, Carla Parra Escartín, Agata Savary, Sascha Bargmann, et al.. PARSEME Survey on MWE Resources. 9th International Conference on Language Resources and Evaluation (LREC 2016), May 2016, Portorož, Slovenia. pp.2299-2306. ⟨hal-01316351⟩