A Hybrid Strategy to Personalize the Digital Television by Semantic Inference

Abstract

The digital TV (DTV) will bring a significant increase in the number of channels and programs available to end users, with many more difficulties for them to find interesting programs among a myriad of irrelevant contents. So, automatic content recommenders should receive special attention in the following years to improve the assistance to users. However, current techniques of content recommenders have important well-known deficiencies, which complicates their wide acceptance. In this paper, a new hybrid approach for automatic TV content recommendation is proposed based on the so-called Semantic Web technologies, that significantly reduces those deficiencies. The strategy uses ontology data structures as a formal representation both for contents and users’ profiles. The approach has been implemented in the AdVAnced Telematics search of Audiovisual contents by semantic Reasoning (AVATAR) tool, a new TV recommender system that makes extensive use of well-known standards, such as TV-Anytime and Web ontology language (OWL). Also, an illustrative example of the kind of reasoning carried out by AVATAR is included, as well as an experimental evaluation of the performance achieved.