An Agent-Based Hierarchical Clustering Approach for e-Commerce Environments

In this paper we propose an agent-based hierarchical clustering technique operating on both user profiles and e-commerce sites. We show, by designing a number of appealing applications, how exploitation of such hierarchies can really benefit both customers and vendors in their activity. As a support for the customer, we provide a categorization of sites of interest as she/he perceives them, and also dynamic ecommerce portal personalization. As a support for the vendor, we design a categorization of customers directly supporting knowledge discovery on customer behavior.