Realizing Adaptive Web Sites Based On Web Mining

As the information of website is more and more abundant as well as the topology more and more complex, there exists the situation of"information overloading"and"resource maze"in information service. From a user's perspective, different users have different access intention and the access intention of one user may vary at different times. On the other hand, only administrators of website satisfy access needs of users in time can improve information service quality and boost more economic performance.Adaptive web sites emerge as the times require to solve the problems puzzles both user and website administrators. It automatically improves the organization and presentation by learing form visitor access patterns mined from web server logs, with intention to produce more easily navigable web sites. Being a hotspot problem of AI (Artificial Intelligence) since 1996, researches on adaptive web sites have made some achievements along with the development of data mining in recent years.Catching up on information that how the website is accessed by users is guarantee of realizing high performance of adaptive web sites. This paper mainly researches on how to analyze web server logs with web mining technology to obtain the information, which are transformed to intelligence of websites to provide individual recommendations or different view to different users, as well as optimize topology of website adaptively. So the research is focused on the following aspects.(1) Analyzes demands of adaptive web site, discusses the user access model and different user access patterns that can be mind with web mining technology.(2) Researches on problems in the process of web log data pretreatment, in order to offer exact fountain for different kinds of mining method.(3) Analyzes advantage and disadvantage of classic mining methods applied in realizing adaptive web sites, brings forward algorithm for individual recommendations based on association rules, sequence rules and clustering respectively according to the localization.(4) Discusses different kinds of guidelines that measure the recommendation system to evaluate adaptive web sites performance. And evaluates algorithms in this paper from the standpoint of theory and experiment based on Chongqing Univercity web site.(5) A prototype system of adaptive web site based on web mining technology,...