For a broadtopic and for a ambiguous query, different users may have distinct search goals when they submit it to the search engine. The analysis and inference of user search goals can be very useful in improving the search engine user experience and relevance. In this article, we propose a novel approach to infer the user search goals by analyzing query logs in search engine. First, we propose a framework to find different user search goals for a single query by gathering the proposed feedback sessions. Feedback sessions are constructed by user clickthrough logs and can be efficiently reflect information needs of the users. Second is, we propose a novel approach to generate pseudodocuments to better represent feedback sessions for gathering. Finally, we propose “Classified Average Precision i.e CAP” to evaluate performance of the inferring user search goals. we included fuzzy search into this article which makes user to query the search engine in easy way. It takes the keyword and searches in the corresponding database. By having this facility efficiency of the system increases
展开▼
