Cross-domain Recommendations based on semantically-enhanced User Web Behavior

Information seeking in the Web can be facilitated by recommender systems that guide the users in a personalized manner to relevant resources in the large space of the possible options in the Web. This work investigates how to model people's Web behavior at multiple sites and learn to predict future preferences, in order to generate relevant cross-domain recommendations. This thesis contributes with novel techniques for building cross-domain recommender systems in an open Web setting.