Study Of Data-Ming Based On WEB

Internet is a huge and global information service center, which provides various kinds of information services. However, with the explosive growth of data available on the web, availability of this information becomes more and more important. Being based on this background, the application of web mining is raised.In this paper, we introduce the basic knowledge of data-mining and the classes of Web Mining. According to web information, there are three types of available: Web Usage Mining, Web Content Mining, and Web Structure Mining. The research direction of this paper is Web Usage Mining. The web usage mining is that mining user's action model to visit web site using data-mining technology and web log data. In this paper, the emphases is the arithmetic's application and mend of the data pretreatment and transaction identification in the web usage mining, and the arithmetic is effective has proven true. Besides, we applied the association rule to web usage mining, and brought forward a mining arithmetic of frequency path in order to finding the visitation model of users of web site.