LessLex: Linking Multilingual Embeddings to SenSe Representations of LEXical Items

2020-06-01 · CL 2020 6 · Davide Colla, Enrico Mensa, Daniele P. Radicioni

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.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Tasks

text similarity

Automatically Linking Lexical Resources with Word Sense Embedding Models

2018-08-01 · COLING 2018 8 · Luis Nieto-Pi{\~n}a, Richard Johansson

Automatically learnt word sense embeddings are developed as an attempt to refine the capabilities of coarse word embeddings. The word sense representations obtained this way are, however, sensitive to underlying corpora …

Word Embeddings

UNIBA: Combining Distributional Semantic Models and Sense Distribution for Multilingual All-Words Sense Disambiguation and Entity Linking

2015-06-01 · SEMEVAL 2015 6 · Pierpaolo Basile, Annalina Caputo, Giovanni Semeraro

All Entity Linking Named Entity Recognition (NER)Word Sense Disambiguation

Semantic Textual Similarity Transfer Learning Word Embeddings

LessLex: Linking Multilingual Embeddings to SenSe Representations of LEXical Items

Code (0)

Tasks

Similar Papers 제목 키워드 기반

Automatically Linking Lexical Resources with Word Sense Embedding Models

UNIBA: Combining Distributional Semantic Models and Sense Distribution for Multilingual All-Words Sense Disambiguation and Entity Linking

Multilingual Word Sense Disambiguation and Entity Linking

SemEval-2015 Task 13: Multilingual All-Words Sense Disambiguation and Entity Linking

Sense representations for Portuguese: experiments with sense embeddings and deep neural language models