With the rising integration of more and more online data sets with various hardware platforms, such as set-top boxes, mobile phones, it becomes increasingly crucial to have intelligent systems guiding the user through the maze of data. A necessity in such systems is a comprehensive user model which allows the system to know how and what the user might expect. In iFanzy, a context-sensitive personalized TV guide, we amass such a user model by monitoring the user's actions and behavior and extensively making use of Semantic Web techniques to interpret these activities and offer in the right moment the right content to the user. We focus on solving the 'cold start' problem by importing user data from external to iFanzy online applications. This is realized by exploiting background ontological knowledge to map the imported concepts, so that iFanzy can minimize the effort in accumulating a complete user model.
|Title of host publication||Proceedings of the 7th European Interactive Television Conference (EuroITV'09, Leuven, Belgium, June 3-5, 2009)|
|Place of Publication||New York NY|
|Publisher||Association for Computing Machinery, Inc|
|Publication status||Published - 2009|