An approach for evaluating User Model Data in an interoperability scenario

Since more and more interactions take place between humans and different user-adapted applications in daily life, there is the great opportunity to acquire a lot of knowledge about a user and use it in order to reach a better adaptation. A common vision states that user model should be interoperable and shareable across user-adaptive applications in different contexts and environments. The core idea of our approach is that the interoperability of user model data lets to reach more effective adaptation results only if the exchanged data are reliable. In this paper we discuss how this goal can be achieved by a semantic annotation of the exchanged data, providing the requestor with the possibility to evaluate i) the reliability of the exchanged value and ii) the reputation of the provider system. To do this, we propose a User-Adapted System Meta-Description which identifies a user-adapted system enriched with a set of semantic meta-information regarding the provider system, the exchanged value, the reasoning strategies used to define the value.