On Clustering Visitors of a Web Site by Behavior and Interests.

This paper addresses the issue of how to define clusters of web visitors with respect to their behavior and supposed interests. We will use the non-obvious user profiles (NOPs) approach defined in [10], and present a new clustering algorithm which is a combination of hierarchical clustering together with a centroid based method with priority, which allows to cluster web users by similar interest in several topics. [ABSTRACT FROM AUTHOR]