Utilizing inter-passage and inter-document similarities for re-ranking search results

We present a novel language-model-based approach to reranking an initially retrieved list so as to improve precision at top ranks. Our model integrates whole-document information with that induced from passages. Specifically, interpassage, inter-document, and query-based similarities are integrated in our model. Empirical evaluation demonstrates the effectiveness of our approach. Categories and Subject Descriptors: H.3.3 [Information Search and Retrieval]: Retrieval models General Terms: Algorithms, Experimentation

• Type: Conference
• Year: 2009
• Where: CIKM
• Authors: Eyal Krikon, Oren Kurland, Michael Bendersky
• Keywords: CIKM 2009, Database, Models General Terms, Query-based Similarities, Subject Descriptors