Indexat a

Llicència i ús

Citacions

Altmetrics

Anàlisi d'autories institucional

Compartir

Word Sense Disambiguation for Linking Domain-Specific Resources

Publicat a:Ceur Workshop Proceedings. 3797 164-172 - 2024-01-01 3797(), DOI:

Autors: Palacios L

Afiliacions

Resum

Word Sense Disambiguation (WSD) is a traditional task in Natural Language Processing (NLP), which involves determining the meaning of words based on their context and differentiating between multiple possible senses. Current approaches, reliant on general-purpose lexical resources like WordNet and Wikidata, often present limitations for domain-specific tasks. This research focuses on addressing the challenge of lexical ambiguity in specialized terminologies in Spanish, aiming to enhance the interconnection and interoperability of those resources. The proposed approach is based on accurately tagging terms with specific senses using automatically integrated sense inventories and hybrid WSD algorithms enabling the use of Entity Linking (EL) and Entity Matching (EM) techniques to facilitate the transformation of disambiguated terminologies into Linked Data formats. Ultimately, this work aims to contribute to the advancement of domain-specific applications, improving semantic analysis and knowledge extraction in specialized fields. © 2024 Copyright for this paper by its authors.

Paraules clau