Handling Semantic Heterogeneity in Interoperable Distributed User Models

Due to the overspread of user adaptive systems user data are collected and processed in diverse settings and from different platforms. The computational effort to extract user models is commonly repeated across applications and domains, mainly due to lack of interoperability and synchronization among user-adaptive systems. One way of achieving a complete picture of a user's experience is to allow systems to share user data to obtain maximum leverage and reuse of information. We address this process as user model interoperability. One of the major challenge to user models interoperability is handling semantic heterogeneity. The paper proposes a new approach for user model interoperability which deals with the semantic heterogeneity of user models and automates the user model exchange across applications.