Paper: Bridging Topic Modeling and Personalized Search

ACL ID C10-2134
Title Bridging Topic Modeling and Personalized Search
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010

This work presents a study to bridge topic modeling and personalized search. A probabilistic topic model is used to extract topics from user search history. These topics can be seen as a roughly summary of user preferences and further treated as feedback within the KL-Divergence re- trieval model to estimate a more accurate query model. The topics more relevant to current query contribute more in updat- ing the query model which helps to dis- tinguish between relevant and irrelevant parts and filter out noise in user search history. We designed task oriented user study and the results show that: (1) The extracted topics can be used to cluster queries according to topics. (2) The pro- posed approach improves ranking qual- ity consistently for queries matching user past interests and is robu...