The Research Of Implementation Methods On The Personalized Information Services

The dissertation brings forward the thought of an implementation method of Personalization Information Assistant (PIA) and a model of Web-based three layers agent framework structure to solve the problem on the basis of analyzing the localization of the technologies of traditional personalization information services and its present status in the area. The functions in the model are interrelation and independent each other, which may be free from increase or reduction by designer, and each technique of every layers can be studied and achieved on its own. The PIA's ability can be endued with the personalized information system. The model is very flexible and operational, and this method has the universality for personalization information services.In accordance with the idea of user-centered design (UCD) process, we consider that the key aspects of research are the extraction approaches of the personal characteristics of the user. By means of Artificial Intelligent Agent techniques and Data Mining techniques, the user interests are automatically learned and discovered. The several approaches have been studied that range from the customizing service, asking relevant questions, the relevant feedback and Web-based usage data mining can find out the interest features of the user and the patterns of behavior, to provide users with what want or need the personalized information services.By training, the PIA learns gradually interests of the individual users and provides users with what need the information, and informs the user while Web pages have been changing in time so as to lighten the burden of users during the process of retrieval information.The practice of the research indicates that the implementation method of the personalized information services can effectively reduce the complexity degree of the information system. The approaches have been developed can discover user interests, and the goal of data manage has been come true in the personalization of information services by PIA.