LessLex

Linking Multilingual Embeddings to SenSe Representations of LEXical Items

Dados Bibliográficos

ID12155757
AutoresDavide Colla (0000-0002-9999-0109, University of Turin - Italy, Computer Science Department., autor correspondente), Enrico Mensa (0000-0001-7743-4999, University of Turin - Italy, Computer Science Department., autor correspondente), Daniele P Radicioni (0000-0003-0443-7720, University of Turin - Italy, Computer Science Department., autor correspondente)
Ano2020
Volume46
Fascículo2
Páginas289-333
Data de publicação2020-03-23
Peer ReviewedSim
Open AccessSim
TipoARTICLE
PeriódicoComputational Linguistics (JOURNAL)
Identificadores do periódicoISSN: 0891-2017 • E-ISSN: 1530-9312
EditoraAssociation for Computational Linguistics (PUBLISHER • US)
DOI10.1162/coli_a_00375
OpenAlexW3013895536
IdiomaEN
Citações recebidas2
Referências citadas81

We present LESSLEX, a novel multilingual lexical resource. Different from the vast majority of existing approaches, we ground our embeddings on a sense inventory made available from the BabelNet semantic network. In this setting, multilingual access is governed by the mapping of terms onto their underlying sense descriptions, such that all vectors co-exist in the same semantic space. As a result, for each term we have thus the “blended” terminological vector along with those describing all senses associated to that term. LESSLEX has been tested on three tasks relevant to lexical semantics: conceptual similarity, contextual similarity, and semantic text similarity. We experimented over the principal data sets for such tasks in their multilingual and crosslingual variants, improving on or closely approaching state-of-the-art results. We conclude by arguing that LESSLEX vectors may be relevant for practical applications and for research on conceptual and lexical access and competence

Information retrieval · Lexical item · Lexical Semantics · Natural language processing · Semantic network · Semantic similarity · Semantic space · Semantics (computer science · Similarity (geometry · Vector space · Advanced Text Analysis Techniques · Computer Science · Mathematics · Natural Language Processing Techniques · Topic Modeling · Artificial Intelligence

Ferramentas

Abrir DOIOpen Access