Indexing with WordNet Synonyms May Improve Retrieval Results

Abstract

Abstract. This paper describes a method developed for the Robust -Word Sense Disambiguation task at CLEF 2009. In our approach, a WordNet expanded index is generated from the disambiguated document collection. This index contains synonyms, hypernyms and holonyms of the disambiguated words contained in documents. Query words are integrated by terms extracted by means of a pseudo relevance feedback technique. The set of terms made of query words and terms resulting from pseudo relevance feedback are searched for in both the expanded WordNet index and the default index. The results show that the use of the extended index did not prove useful, obtaining 14 − 16% less in MAP with respect to the base system. However, for some queries, expanding index terms with synonyms resulted particularly useful.