Abstract

In this paper, we present an approach to reduce the effort of building suitable user models and to ease the administration of cross-application data management. We introduce a centralized sophisticated multiapplication user modeling and profile management framework with an integrated central identity management, called User Modeling and Personalization Framework (UMP). The UMP enables applications to collect and organize their data and extend their user knowledge through data inferences. The data is collected on one central server and the applications can share data among each other. Based on the data collected in different profiles, the UMP builds ubiquitous user models, which are the backbone for understanding the users’ interests and demands. Based on the computed user models, adaptive software facilitates interaction for the user, e.g. by highlighting important functionality or omitting unnecessary information.