Does word sense disambiguation improve information retrieval?

Abstract

A basic form of semantic annotation is to label a word in a document with its correct sense based on the context in which the word occurs, thus providing the disambiguated sense of the word. Performing this task automatically is known as word sense disambiguation, which has been extensively studied in the natural language processing literature. Will semantic annotation of word senses improve information retrieval? This paper provides some thoughts on this question, which lies at the intersection of natural language processing and information retrieval. © 2011 Author.