An approach to clustering Web browsing patterns by ART2 neural networks with general learning rules

Categorising visitors based on their interaction with a website is a key problem in Web content usage. The purpose of clustering users based on users' access patterns in a particular website is to find groups of users with similar interests and motivations for visiting that website. The clickstreams generated by various users often follow distinct patterns, the knowledge of which may help in providing customised content. This paper proposes a novel approach for weblog clustering based on ART2 neural networks with generalised learning. An advantage of the proposed approach is that it can gradually "forget" poorly populated clusters, thus releasing network resources for future use. Such approach could be applied as an efficient weblog analysis tool particularly useful for Web sites with huge number of logged clickstreams or rapidly changing Web content.