The Research Of Adaptive Web Sites Based On Web Mining

With the swift and violent development of Internet technology and e-commerce, Web is dramatically changing our lives unprecedented. Because more business transactions and servies are carried out through the Web, better services for the need of Web-based applications and understanding the action of customers become the focus of attention today. In order to solve the problems of relationship between customers and providers, adaptive Web sites become to the focus of study at present.Logs of user accesses to a site provide an opportunity to observe users interacting with that site. Through web usage mining this paper aims to research and build the adaptive web sites.This article aims to provide a comprehensive research on the principles and algorithm of building adaptive sites. Through analyzing the types of websites and users' browsing habits, it proposes a fully new session identification algorithm and trail path complementary algorithm for tree sites containing pop_up pages, and also provides the improved maximum forward frequent trail path algorithm and object page association algorithm.To practice and verify the proposed algorithm and realize the J2EE based adaptive site MAWSS, which consists of four modules: data pretreatment, site adaptation, page recommendation and object page association. Among the four modules, the data pretreatment and the site adaptation play a basic and central role respectively.