Adaptive document clustering using incrementally expanded queries

This paper describes a method of increasing the effectiveness of browsing by clustering large numbers of information retrieval results, and discusses the implementation of adaptive document clustering. In the proposed method, query information is successively expanded and refined in the course of browsing, and it is utilized as a clustering viewpoint that expresses the interests of the user. The proposed method is experimentally evaluated in an application to WWW information retrieval and its effectiveness is demonstrated. © 2001 Scripta Technica, Syst Comp Jpn, 32(2): 64–74, 2001