Intent-aware search result diversification

• DOI: 10.1145/2009916.2009997
• Publication Date: 2011-07-26T14:09:37Z
• Authors: Rodrygo L.T. Santos; Craig Macdonald; Iadh Ounis
• Keywords: Relevance

Search result diversification has gained momentum as a way to tackle ambiguous queries. An effective approach this problem is explicitly model the possible aspects underlying query, in order maximise estimated relevance of retrieved documents with respect different aspects. However, such themselves may represent information needs rather distinct intents (e.g., informational or navigational). Hence, diverse ranking could benefit from applying intent-aware retrieval models when estimating In paper, we propose diversify results for given by learning appropriateness each query. Thorough experiments within evaluation framework provided diversity task TREC 2009 and 2010 Web tracks show that proposed can significantly improve state-of-the-art approaches.