The effect of similarity measures on the quality of query clusters.

Abstract

Query clustering is a process that can be used to discover common interests of online information seekers and to exploit their collective search experience for the benefit of others. Harnessing such search experiences facilitates collaborative querying that in turn may help users of digital libraries and other information systems to better meet their information needs. Since similarity is fundamental to the definition of a cluster, measures of similarity between two queries is essential to the query clustering procedure. In this paper, we examine the effectiveness of different similarity measures. A set of experiments was carried out to study the impact of different similarity measures on the quality of query clusters. The results show that different similarity measures outperform each other in different query cluster quality criteria. Implications for these findings are discussed.