Based On The User's Personalized Search Engine

Along with the rapid development of the Internet around the world, more and more information has been shown in the Internet. Users of SE face the problem of resource overloading and can not easily find useful information in the worldwide Internet, therefore, personalized search engine become a valid way to advance the situation or settle the problem. Personalized search engine and personalized technology become hot research subjects, and some research achievements come to existence.The paper presents the status of the personalized search engine and the deficiencies of personalized search engines. By making the user model and analyzing the user action in the course of using search engine, a new full research framework of personalized search engine called Personalized Search Engine based on user is proposed. It mainly takes charge of information collection of user action and user feedback technology. The research is of value in both theory and application. The main work in the paper is showed as follows:Firstly, the research tells basic theory of personalized search engine, and the method, contents and storage of collection of user information based on making user model. According to the theory above, it presents the ways of making user model based on user description files.Secondly, through collecting user information, the thesis presents a kind of method about renewing user model based on this technology. Further, to get united to the characters of user model, it tells a new sort of personalized page ranking—personalized page ranking on the basis of user cache. In the end, according to the characters of personalized search engine, it describes two kinds of performance index about evaluating personalized search engine, user action index and user action precision.Finally, an experimental system called PSE is established and proves the feasibility of architecture about personalized search engine. Besides, through collecting user information, a new user model is set up, in some percents, and it improves query precision of inner search engine. According two performance indexes about average user action index and average user precision, the experiment demonstrate that ranking method based on user cache has certain advantages. It also shows two evaluation indicators in the performance evaluation of personalized search engine are consistency.