Topic Exploration and Distillation for Web Search by a Generalized Similarity Analysis

Topic distillation is the process of finding representative pages relevant to a given query. Well-known topic distillation approaches such as the HITS algorithm have shown to be useful in identifying high quality pages of the most popular topic within a query specific graph of hyperlinked documents. Many succeeding researchers focus on augmenting HITS with further content analysis to alleviate the steady deterioration of distillation quality suffered by HITS. In this paper, we attempt to revisit the behavior of HITS from a different point of view. Namely, a similarity-based analysis model is applied to observing the distillation procedure. By defining a generalized similarity, an algorithm is proposed, which can improve the quality of distillation only using the information of hyperlinks. A topic exploration function is also integrated in the algorithm framework, which enables end-users to search less popular topics when multi-topics are involved in queries. The experimental results reveal two benefits from the new algorithm: the improvement of distillation quality without utilizing any content information of pages, and an additional ability to explore the topics emerging in the query results.