Promoting ranking diversity for biomedical information retrieval using wikipedia

Abstract

In this paper, we propose a cost-based re-ranking method to promote ranking diversity for biomedical information retrieval. The proposed method concerns with finding passages that cover many different aspects of a query topic. First, aspects covered by retrieved passages are detected and explicitly presented by Wikipedia concepts. Then, an aspect filter based on a two-stage model is introduced. It ranks the detected aspects in decreasing order of the probability that an aspect is generated by the query. Finally, retrieved passages are re-ranked using the proposed cost-based re-ranking method which ranks a passage according to the number of new aspects covered by the passage and the query-relevance of aspects covered by the passage. A series of experiments conducted on the TREC 2006 and 2007 Genomics collections demonstrate the effectiveness of the proposed method in promoting ranking diversity for biomedical information retrieval. © 2010 Springer-Verlag Berlin Heidelberg.