Agent Technologies for the Development of Adaptive Web Stores

With the increasing popularity of Web shopping, the personalization of the front-end of on-line stores has become a critical issue: these systems are accessed by customers with different backgrounds, expertise and preferences; therefore, their usability can only be improved by personalizing their interfaces to the needs of each specific user. In this paper, we describe the architecture of SETA, a prototype toolkit for the creation of adaptive Web stores, where user modeling and flexible hypermedia techniques are exploited to personalize the interaction with the user, dynamically generating the catalog pages on the basis of the user's features. The SETA architecture is composed by different specialized agents, which cooperate to the management of the interaction with the user; we will discuss how the application of agent-based techniques has been essential to the design and development of this system.