An Efficient Approach for Retrieving Personalized User Goals

The most important facility of a search engine is to retrieve relevant information as early as possible. For this understanding user search goal is necessary. For the same query the user information need may be different. In this paper we propose a method to infer user search goal by analyzing search engine query logs. User search goals are discovered by clustering proposed feedback sessions. Feedback sessions are constructed from user click-through logs and it reflect information need of users. Then propose novel approach to generate pseudodocuments which represent feedback session for clustering. We present a large-scale evaluation framework for personalized search based on query logs and then evaluate five personalized search algorithms (including two click-based ones and three topical-interest-based ones). It represents a significant improvement over generic Web search for some queries, while it has little effect and even harms query performance under some situations.