A study on user profile reasoning for target advertisement in a TV personalization framework

Abstract

The traditional broadcasting services such as terrestrial, satellite and cable broadcasting have been unidirectional mass media regardless of TV viewer``s preferences. Recently rich media streaming has become possible via the broadband networks. Furthermore, since bidirectional communication is possible, personalcasting such as personalized TV services has been an emerging technology by taking into account the user``s preference on content genres, viewing times, and actors/actresses, etc. Accordingly, personal media becomes an important means for content provision service in addition to the traditional broadcasting service as mass media. The personal media means contents considering individual preferences. In order to provide media services in personalized ways, TV personalization framework and its related technologies are essential. In the thesis, we propose a TV personalization framework which includes pEPG (personalized Electronic Program Guide) generation based on user preferences, target advertisement service, and content mobility service for consuming contents in some personalized ways. In the thesis, a user profile reasoning method for target advertisement is developed in the TV personalization framework which supports the content mobility service for multi-users. The user profile reasoning method predicts unknown TV viewer``s gender and ages by analyzing TV program viewing history. Based on the estimated user``s gender and ages, targeted advertisement becomes possible. The target advertisement is developed based on such a user profile reasoning algorithm. In the proposed TV personalization framework, pEPG can be generated based on the user preference values that are calculated by the user preference learning algorithm. Users can browse, select, and consume their preferred TV program contents from the given pEPG. Also, the content mobility is possible based on user preference via various kinds of user terminals in the TV personalizatio...