Sense Vocabulary Compression through the Semantic Knowledge of WordNet for Neural Word Sense Disambiguation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Sense Vocabulary Compression through the Semantic Knowledge of WordNet for Neural Word Sense Disambiguation

Résumé

In this article, we tackle the issue of the limited quantity of manually sense annotated corpora for the task of word sense disambiguation, by exploiting the semantic relationships between senses such as synonymy, hypernymy and hyponymy, in order to compress the sense vocabulary of Princeton WordNet, and thus reduce the number of different sense tags that must be observed to disambiguate all words of the lexical database. We propose two different methods that greatly reduces the size of neural WSD models, with the benefit of improving their coverage without additional training data, and without impacting their precision. In addition to our method, we present a new WSD system which relies on pre-trained BERT word vectors in order to achieve results that significantly outperform the state of the art on all WSD evaluation tasks.
Fichier principal
Vignette du fichier
GWC_2019.pdf (295.46 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02131872 , version 1 (16-05-2019)

Identifiants

  • HAL Id : hal-02131872 , version 1

Citer

Loïc Vial, Benjamin Lecouteux, Didier Schwab. Sense Vocabulary Compression through the Semantic Knowledge of WordNet for Neural Word Sense Disambiguation. Global Wordnet Conference, 2019, Wroclaw, Poland. ⟨hal-02131872⟩
142 Consultations
227 Téléchargements

Partager

Gmail Facebook X LinkedIn More