Mining user browsing pattern based on representative set

Abstract

Web log contains large volume of user browsing information. How to mine user browsing patterns is an important research topic. On the analysis of the present algorithms for mining user interests, user-interest can be used for mining user browsing patterns. In order to cluster overlapping objects effectively, a new concept-Cut Plane is introduced. And for achieving reasonable and acceptable running speed, representative set is presented too. On all of these foundations, RSBUBP (Representative Set Based User Browsing Patterns) algorithm is proposed. Experiments showed that it was accurate and scalable. It's suitable for application in E-business, such as to optimize Web site or to design personalized service.