Collaborative filtering for sharing the concept based user profiles

User profiling strategy is an essential and fundamental component in search engine personalization. Recent research has focused on the automatic learning of user preferences from users search histories or browsed documents and the development of personalized systems based on the learned preferences. In this paper we focus on developing three concepts based user profiling methods that are based on users both positive and negative preferences. Also an agent based approach to collaborative filtering is applied, where agents work on behalf of their users to form shared interesting groups. This pre-clustering process allows users with the same interest to share their profiles. These shared profiles are dynamically updated to reflect the users evolving interest over time. The performance with the changing user interest of user is evaluated as experimental results. We noticed what would happen to the performance that of other collaborative filtering methods and also measure its performance in different domains.