Promoting Ranking Diversity for Biomedical Information Retrieval Using Wikipedia

Traditional Information Retrieval models assume that the relevance of a document is independent of the relevance of other documents. However, in reality, this assumption may not hold. The usefulness of retrieving a document usually depends on previous ranked documents, since a user may want to see the top ranked documents concerning different aspects of his/her information need instead of reading relevant documents that only deliver redundant information. In this talk, I will discuss how to find relevant documents that can deliver more different aspects of a query. In particular, I will discuss new models derived from the survival analysis theory for measuring aspect novelty. I will discuss how to use Wikipedia to detect aspects covered by retrieved documents. An aspect filter based on a two-stage model will be introduced and a new re-ranking method that combines the novelty and the relevance of a retrieved document at the aspect level will also be presented. Through extensive experiments on standard large-scale TREC biomedical collections, I will show that the proposed models and methods are effective in promoting ranking diversity for biomedical information retrieval.