Abstract

Query expansion is one of the important research topics in information retrieval system. A novel method for query expansion is presented in this paper. The proposed method is a hybrid QE technology that combines user relevance feedback and ontologies. The proposed method has two significant novelties: it combines user relevance feedback with ontology and uses FirteX as the experimental platform. We compared the proposed method with cosine similarity-based QE which is a widely used query expansion technique. The experimental results show that the proposed method outperforms cosine similarity-based QE 15% and 13% in terms of average precision and average recall and has an improvement of 16% in F-measure.

Paper Details