Research Of Path Clustering Based On The Access Interest Of Users

Internet is changing our life gradually, which becomes the main way of transmitting information throughout the whole world. The future is the world of web. How to attract more users and improve the browse interest of the users become the major work of web sites. At the same time, it raises higher claim to the design and function of the web sites. Web sites should have intelligence, which can find out information what the users need quickly and accurately, can provide different services for different users, can allow users to design pages according to their own need, can provide the information of product sale strategy for the users, and so on.From the view of managers of web site, they need better automatic assistant design tool. They can adjust the page structure automatically according to users' access interest, access frequency and access time. They can improve their service and develop purposeful e-commerce to satisfy the visitors' need. Anefficient way to solve the problem is to make use of the technique of web data mining to get the useful information.Web sites classify users with the same interested making use of clustering technology by attending access path of users, access time of one's page, the staying times of at the page and URL of linking to the page.The paper will introduce a new clustering arithmetic of basing on access interested of users. It defines new interest degree, similitude degree, and clustering center.The arithmetic is used to clustering the path of users' access web site. Compared with other arithmetic, k-pathsplus arithmetic pays more attention to users' access interest by the path of users' access. Therefore the designer of web site can adjust the page structure design and physics storage design according to clustering results, at the same time it can lighten users' burden by avoiding users' input, and clustering results can be used for instant individual users' access interface. In the end we do a true experiment making use ofwww.ty.sx.cn log file. The result of experiment proves our arithmetic is successful.The paper makes a summary in the end, and at the same time it puts forward some prospects for research work.