Inferring User Search goals Engine Using Bisecting Algorithm
Deepali Agale, Beena Khade · 2014
Different users may have different search goals when they submit broad-topic and ambiguous query, to a search engine. The inference and analysis of user search goals can be very useful in improving performance of search engine. To infer user search goals by analyzing search engine que- ry logs a novel approach is proposed. First, we propose a framework to find out different user search goals for a query by clustering the proposed feed- back sessions. Feedback sessions are built from user click-through data and can efficiently reflect the information needs of users. Second, then propose a novel approach to generate pseudo-documents by using feedback sessions for clustering. For clustering we use a new algorithm which is bisecting K- means algorithm. At the end, a new criterion Classified Average Precision (CAP) is proposed to evaluate the performance of inferring user search goals.