Ant Colony Algorithm for Arabic Word Sense Disambiguation through English lexical information

Abstract : The ability to identify the intended meanings of words in context is a central research topic in natural language. Many solutions exist for word sense disambiguation (WSD) in different languages, such as English or French, but research on Arabic WSD remains limited. The main bottleneck is the lack of resources. In this article, we show that it is possible to build a WSD system for the Arabic language thanks to the Arabic WordNet and its connexions to the English Princeton WordNet. Given that the Arabic WordNet does not contain definitions for synsets, we construct a dictionary that maps the Princeton WordNet definitions to the Arabic WordNet. We also create an Arabic evaluation corpus and gold standard. We then exploit this dictionary and evaluation corpus to run and evaluate an adapted Ant Colony algorithm on Arabic text that can use the Lesk similarity measure thanks to definition mapping. The algorithm shows a performance of approximately 80% compared to the random baseline of 78.9 %.
Type de document :
Article dans une revue
International Journal of Metadata, Semantics and Ontologies, Inderscience, 2015
Liste complète des métadonnées

Littérature citée [38 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01684576
Contributeur : Didier Schwab <>
Soumis le : lundi 15 janvier 2018 - 15:53:33
Dernière modification le : jeudi 11 octobre 2018 - 08:48:03
Document(s) archivé(s) le : dimanche 6 mai 2018 - 02:57:37

Fichier

authorFinalVersion.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01684576, version 1

Collections

Citation

Bakhouche Abdelaali, Yamina Tlili-Guiassa, Didier Schwab, Andon Tchechmedjiev. Ant Colony Algorithm for Arabic Word Sense Disambiguation through English lexical information. International Journal of Metadata, Semantics and Ontologies, Inderscience, 2015. 〈hal-01684576〉

Partager

Métriques

Consultations de la notice

76

Téléchargements de fichiers

153