Intelligent Information Retrieval Personalized System Based On Agent

With the rapid development of the Internet, people can get the required information more easily and directly. Nowadays, we have many Internet search engines, such as Google, Yahoo, and WebCrawler, which can help people search the Internet for their requirements. However, there are many shortcomings. Due to the ambiguity and polysemy of words, the engine can not express users'intention or meaning precisely. It can not make distinctions among users. Moreover, it can not find and collect the required information automatically; if the user wants to search the same topic to get updated information, he has to search it again which is a waste of time.Facing the current situation of network information service, people desire a kind of service model that the required information for users can be recommended to them actively. That is the personalized active information service. The intelligent agent technology plays an important part in carrying on the personalized active information service.To solve the problem of the insufficiency of information retrieval system, this paper has firstly systematically studied the key technology dealing with the Agent-based personalized active information service and puts forward an Internet information retrieval system model for the agent. The characteristics of personalized information retrieval are discussed, and a framework of person-alizing information retrieval system is presented.The system consists of three parts: User Agent, Information Search Agent, and Information Filtering Agent. We research the three models separately. The User Agent is the key research topic in this paper. We should form and train the User Agent through the expression of information requirement and information feedback by the user. A creating and a modifying algorithm of the users'interests based on this model and a method of using and creating a tree of user's interests to analyze the user's search quests are presented subsequently. The information search Agent not only realizes meta-search, but also makes self-search on the web through its inquiry agent connecting to Internet search engines when the recalls can not meet the needs of the user. The search algorithm starts its searching from the recalls of the inquiry agent to reduce the range of searching and to increase the speed of searching. According to the user ready-made information resources, the informa-tion filtering Agent analyses the user's favorites and adopts vector space method to carry out the personalized information filter.Finally, in conclusion, the paper gives a brief discussion on the future research of personalizing information retrieval.