The User Interests Model With Real-time Updated User Interests Based On User Behaviors And Similar Semantic

Personalization Services System is developed to help people obtain the just needed information from large-scale information collections. In the past, people placed importance on the technology of personalization services, such as recommendation technology, information searches technology, user clustering technology and so on. Then the importance of user models in personalization services has been neglected. So in this paper, the research on user models is most important. In Chinese, character compartmentation is very important in user interests models, but it is often neglected, too. Now there isn't a good enough method to solve the problem of character compartmentation. In this paper, a method to solve the problem of character compartmentation is proposed.The user interests models in the existing Recommender Systems can express the user's interests effectively, but they can't modify the user interests models while the user interests change, and cannot adapt the changes in time. This flaw of the existing recommender systems results in the suggestions to users aren't exact. So we propose a recommender system based on user behaviors and similar semantic. The system can adapt the changes of user interests quickly, and is more exact than the existing recommender systems.