Analyzing user profiles in electronic markets

Abstract

The electronic markets have reached a huge volume of users and transactions with the popularization of the Web applications. Thus, it is essential to understand clearly the behavior of the users of these applications to improve the services and also to offer new features to them. In this context, this paper presents a workload characterization and evaluates the navigation profiles of humans and bots in an electronic market. We use a graph model to represent the user navigation, identifying typical differences among their behaviors, as the action of data collecting performed by bots, and the typical act of searching from humans. Besides proposing a methodology for user session tracking characterization, the results of this paper can be applied in order to design new personalization techniques or to improve the user-system interaction.