Context-Awareness for IPTV Services Personalization

Marek Dabrowski, Justyna Gromada, Hassnaa Moustafa
2012 2012 Sixth International Conference on Innovative Mobile and Internet Services in Ubiquitous Computing  
TV and video services landscape is currently undergoing significant changes. Traditional TV broadcasting model is supplemented and often even replaced by digital content distribution services over the Internet. As a result of this trend, a range of services and contents available for users is rapidly expanding. As a side-effect, designing efficient user interfaces for discovering the content, as well as for manipulating associated interactive services, becomes more and more cumbersome.
more » ... computing and context-awareness principles seem to be promising for making user interaction with the system more seamless and fluid. This paper discusses context-awareness as a new trend for services personalization, defining the meaning of "context" and different contextual data relevant for IPTV (Internet Protocol Television) services. A novel architecture for unified storage and processing situational data in IPTV service domain is presented, together with discussion of its implementation issues and validation by testbed experiments. The UP-TO-US context-aware architecture is designed for seamless monitoring of user's environment (including networks and terminals), interpreting user's requirements and making the user's interaction with the TV system dynamic and transparent. Consequently, content personalization is achieved, matching the user's needs and current state of his surrounding environment. † This work has been done by the author during her stay in Orange Labs 49 A context-aware architecture Dabrowski, Gromada, Moustafa and Forestier and preferences. Vast literature is available on automatic recommendation systems (see e.g. quite recent results of real-life experiments with TV recommender systems presented in [10] ). Recommender systems may analyze user's previous content choices to learn and match his habits, or may exploit the similarities between certain user categories to infer their likes and dislikes based on observation of other similar users. Going further, context-aware systems aim to adapt automatically to user's situation, e.g. a contextaware IPTV could find the program that suits the user and his current situation: staying at home or on the train, in the noon or in the evening, being in front of TV set or mobile device. In each situation, the expected behavior and service choice of the user may be different and one can expect that some similarities and patterns could be found and exploited by the recommendation algorithm, e.g. knowing that a user particularly likes watching sports in the evening time, while staying in his bedroom.
doi:10.1109/imis.2012.97 dblp:conf/imis/DabrowskiGM12 fatcat:m74newxjfnhehizzbtnf75hn7e