Edition 1.2 of the PARSEME Shared Task on Semi-supervised Identification of Verbal Multiword Expressions

Carlos Ramisch (1) , Agata Savary (2) , Bruno Guillaume (3) , Jakub Waszczuk (4) , Marie Candito (5) , Ashwini Vaidya (6) , Verginica Barbu Mititelu (7) , Archna Bhatia (8) , Uxoa Iñurrieta (9) , Voula Giouli (10) , Tunga Güngör (11) , Menghan Jiang Polyu (12) , Timm Lichte (13) , Chaya Liebeskind (14) , Johanna Monti (15) , Renata Ramisch (16) , Sara Stymne (17) , Abigail Walsh (18) , Hongzhi Xu (19)

1 TALEP - Traitement Automatique du Langage Ecrit et Parlé
2 BDTLN - Bases de données et traitement des langues naturelles
3 SEMAGRAMME - Semantic Analysis of Natural Language
4 Heinrich Heine Universität Düsseldorf = Heinrich Heine University [Düsseldorf]
5 LLF - UMR7110 - Laboratoire de Linguistique Formelle
6 IIT Delhi - Indian Institute of Technology Delhi
7 Romanian Academy
8 IHMC - Florida Institute for Human and Machine Cognition [Pensacola]
9 UPV / EHU - University of the Basque Country = Euskal Herriko Unibertsitatea
10 ATHENA - Research and Innovation Center in Information, Communication and Knowledge Technologies
11 Boǧaziçi üniversitesi = Boğaziçi University [Istanbul]
12 POLYU - The Hong Kong Polytechnic University [Hong Kong]
13 University of Tübingen
14 JCT - Jerusalem College of Technology
15 UniOr - Università di Napoli L'Orientale = University of Naples
16 NILC, UFSCar
17 Uppsala University
18 DCU - Dublin City University [Dublin]
19 Shanghai International Studies University

We present edition 1.2 of the PARSEME shared task on identification of verbal multiword expressions (VMWEs). Lessons learned from previous editions indicate that VMWEs have low ambiguity, and that the major challenge lies in identifying test instances never seen in the training data. Therefore, this edition focuses on unseen VMWEs. We have split annotated corpora so that the test corpora contain around 300 unseen VMWEs, and we provide non-annotated raw corpora to be used by complementary discovery methods. We released annotated and raw corpora in 14 languages, and this semi-supervised challenge attracted 7 teams who submitted 9 system results. This paper describes the effort of corpus creation, the task design, and the results obtained by the participating systems, especially their performance on unseen expressions.

Domaines: Informatique et langage [cs.CL]

Carlos Ramisch, Agata Savary, Bruno Guillaume, Jakub Waszczuk, Marie Candito, et al.. Edition 1.2 of the PARSEME Shared Task on Semi-supervised Identification of Verbal Multiword Expressions. Joint Workshop on Multiword Expressions and Electronic Lexicons (MWE-LEX 2020), 2020, Barcelona, Spain. ⟨hal-03014927⟩