Parsimonious relevance models

We describe a method for applying parsimonious language models to re-estimate the term probabilities assigned by relevance models. We apply our method to six topic sets from test collections in five different genres. Our parsimonious relevance models (i) improve retrieval effectiveness in terms of MAP on all collections, (ii) significantly outperform their non-parsimonious counterparts on most measures, and (iii) have a precision enhancing effect, unlike other blind relevance feedback methods. Categories and Subject Descriptors H.3 [Information Storage and Retrieval]: H.3.3 Information Search and Retrieval General Terms Algorithms, Theory, Experimentation, Measurement Keywords Parsimonious Models, Language Models, Relevance Feedback

• Type: Journal
• Year: 2008
• Where: SIGIR
• Authors: Edgar Meij, Wouter Weerkamp, Krisztian Balog, Maarten de Rijke
• Keywords: Information Technology, Parsimonious Language Models, Parsimonious Relevance Models, Relevance Models, SIGIR 2008