Bilkent University
Department of Computer Engineering


Cascaded Cross Entropy-Based Search Result Diversification


Bilge Köroğlu
MSc Student
Computer Engineering Department
Bilkent University

Search engines are used to seek information on the web. Retrieving relevant documents for ambiguous queries based on query-document similarity does not satisfy the users because such queries have more than one different meaning. In this study, a new method, cascaded cross entropy-based search result diversification (CCED), is proposed to list the web pages corresponding to different meanings of the query in higher rank positions. It combines modified reciprocal rank and cross entropy measures to balance the trade-off between query-document relevancy and diversity among the retrieved documents. We use the Latent Dirichlet Allocation (LDA) algorithm to compute relevancy scores of the documents to each meaning of the query. The number of different meanings of an ambiguous query is estimated by complete-link clustering. In this study, we also construct the first Turkish test collection for result diversification, BILDIV-2012. The performance of CCED is compared with Maximum Marginal Relevance (MMR) and IA-Select algorithms. In this comparison, the Ambient, TREC Diversity Track, and BILDIV-2012 test collections are used. The results indicate that CCED is the most successful method in terms of satisfying the users interested in different meanings of the query in higher rank positions of the result list.

Keywords: Ambiguous Query, Cross Entropy, Latent Dirichlet Allocation (LDA), Reciprocal Rank, Search Engine, Search Result Diversification (SRD), Test Collection, TREC Diversity Track.


DATE: 03 September, 2012, Monday @ 9:40