Research On Web-Based User Access Information Mining

Internet penetration, the mass of Web information for data mining provides a wealth of resources, Web users to access information on Web-based mining is the most extensive application. Its purpose is to better understand Web-based applications in order to provide quality services. Because of its potential value, Web access to information mining in the area of research and commercial areas have achieved rapid development.On the basis of Data Mining and Web Mining's knowledge, this paper focuses on Web Log Mining. First, It introduced five basic process of Web log data clean-up, the user identification, conversation recognition, and added path Services to identify, based on the cookie to the user identification technology and based on the maximum before the Panel to invoke the recognition method.Second, It introduced the theory of the mining association rules and Apriori algorithm, analyzed the shortcomings of the mining association rules, proposed the mining association rules based on the pages of interesting, which combined the users interesting and the mining association rules, and applied personalized service.Then, by the use of cluster analysis in real-time personalized recommendations of the way through the order to visit a cluster of users and users for every type of the corresponding page recommendation sets, recommended the use of the current engine for users to visit classification, Attributed to the current users of a particular cluster focus, then the current visit of the user pages, users type their users recommend the recommendation of pages set.Finally, build models of the Web log mining technologies in distance education site. In the distance education process, by introducing a Web log mining, not only for the learners to provide personalized learning, but also conducive to long-distance teaching of management.