TREC-7 Experiments: Query Expansion Method Based on Word Contribution.
Keiichiro Hoashi, Kazunori Matsumoto, Naomi Inoue, Kazuo Hashimoto · 1998
1 Introduction This is KDD R&D Laboratories ' first participation in TREC. In this participation, we focused on ex- periments on a novel method of query expansion. The query expansion method described in this paper is based on a measure we call "word con- tribution". Word contribution is a measure which expresses the influence of a word to the similar- ity between the query and a document. From our data, we figured that words which have highly neg- ative contribution can be considered as to being expressive of the characteristics of the data (query or document) in which they exist. We proposed ex- tracting such words from documents highly similar to a query, and adding them to the original query to generate an expanded query. We made exper- iments to evaluate this method, and reported the results in this paper.