Towards the integration of diverse context into advertisement recommendation on mobile devices

IP Television (IPTV) has created a new arena for digital advertising that has not been explored to its full potential. IPTV allows users to retrieve both on demand content as well as recommended content; however, limited research has been applied in the domain of advertising in IPTV systems. The introduction of IPTV, smart devices, and social networks led to the ability to gather a wider set of rich contextual information from multiple diverse sources. Such context could be used to enrich the decision making process of recommending advertisements to tailor better ads for users. This work attempts to create a targeted advertisement system that starts off by gathering a rich set of context information, and then provides viewers with recommended advertisements based on their current detected contextual information. In this paper, we describe our methodology, as well as our experiments designed to evaluate the proposed system. Our experimental results demonstrate both efficiency, as well as enhanced accuracy in the recommendation of context-sensitive advertisements.