Collaborative Ranking Function Training for Web Search Personalization

In this paper, we present a framework for improving the ranking function training and the re-ranking process for web search personalization. Our method is based on utilizing clickthrough data from several users in order to create multiple ranking functions that correspond to different topic areas. Those ranking functions are combined each time a user poses a new query in order to produce a new ranking, taking into account the similarity of the query with each of the topic areas mentioned before. We compare our method with the traditional approaches of training one ranking function per user, or per group of users and we show preliminary experimental results.