Abstract: For a broad-topic and ambiguous query, different users may have different search goals when they submit it to a search engine. The inference and analysis of user search goals can be very useful in improving search engine relevance and user experience. The project proposes a novel approach to infer user search goals by analyzing search engine query logs. First, it proposes a framework to discover different user search goals for a query by clustering the proposed feedback sessions. The feedback session is defined as the series of both clicked and unclicked URLs and ends with the last URL that was clicked in a session from user click-through logs. Second, the pseudo-documents are produced to better represent the feedback sessions for clustering.

The pseudo-documents are clustered using Fuzzy C Means, the fuzzy similarity based self- constructing algorithm. A novel optimization method is used to map feedback sessions to pseudo-documents which can efficiently reflect user information needs and finally, a new criterion   “Classified Average Precision (CAP)” is used to evaluate the performance of inferring user search goals. Experimental results are presented using user click-through logs from a commercial search engine to validate the effectiveness.


Keywords: User Search Goal, Feedback Session, Fuzzy C Means Algorithm