An associative method for Lesk-based word sense disambiguation

One of the most important current problems in natural language processing is word sense disambiguation (WSD). WSD consists of identifying the correct sense of the words in a given text. In this work, we present a novel method for automatic WSD based on the simplified -Lesk algorithm. The proposed me...

ver descrição completa

Detalhes bibliográficos
Autores: Sulema Torres-Ramos, Israel Román-Godínez, E. Gerardo Mendizabal-Ruiz
Tipo de documento: artigo
Estado:Versão publicada
Data de publicação:2017
País:México
Recursos:Universidad de Guadalajara
Repositório:Redalyc-UDG
OAI Identifier:oai:redalyc.org:157052587007
Acesso em linha:https://www.redalyc.org/articulo.oa?id=157052587007
Access Level:Acceso aberto
Palavra-chave:Lengua y Literatura
Alpha
simplified
Lesk algorithm
associative memories
Beta associative memories
Descrição
Resumo:One of the most important current problems in natural language processing is word sense disambiguation (WSD). WSD consists of identifying the correct sense of the words in a given text. In this work, we present a novel method for automatic WSD based on the simplified -Lesk algorithm. The proposed method employs Alpha -Beta associative memories for the relatedness computation between the senses of the ambiguous words and its context. The performance of this method wa s evaluated in terms of precision, recall, and F -score, using the semantically annotated corpora Senseval -2, Semcor, and Semeval -2007. The results show the advantages of the proposed method compared with other Lesk -based state -of-the -art methods.