Web user clustering from access log using belief function

Abstract

In this work, we present a novel approach to clustering Web site users into different groups and generating common user profiles. These profiles can be used to make recommendations, personalize Web sites, and for other uses such as targeting users for advertising. By using the concept of mass distribution in Dempster-Shafer's theory, the belief function similarity measure in our algorithm adds to the clustering task the ability to capture the uncertainty among Web user's navigation behavior. Our algorithm is relatively simple to use and gives comparable results to other approaches reported in the literature of web mining.