Analysis of Web User Clustering based on Users’ Access Behavior

Abstract

World Wide Web overwhelms us with the immense amounts of widely distributed interconnected, rich and dynamic information. Provision of services to users correctly according to their needs is one of the most important issues in Web. However, provision of services to individual users’ need is time consuming and overburden for the web site developers or administrator. Not only for the developers but also for the users, group-based service provision can fulfill this situation at the same time. In this paper, clustering algorithms: Self Organizing Map (SOM) and K-Means are used to analyze the users’ access behavior. The correctness of the clustering algorithms is tested with two external validation indexes. Our implementation results show that SOM gives better results than K-Means.