Context-Aware Ontological Hybrid Recommender System For IPTV

With the huge growing amount of information continuously produced, shared and available online, finding relevant and beneficial contents or services at a single or few clicks have become almost impossible. Most of the time, we will be returned with thousands of irrelevant web links. As such, a recommender system which recommends contents or services that likely meet the user's needs is crucial, especially in the IPTV domain when the choices for program selection has no time and physical boundary restriction. The two major recommendation techniques are content based and collaborative filtering. Nevertheless, such techniques still suffer from several problems such as cold start, data sparsity and over specialization. Our proposed system namely COHRS is a context-aware recommender system based on ontological profiling under the IPTV domain. Ontological approach improves user profiling process and thus improving the accuracy of a recommendation system. Experimental evaluations indicate that COHRS is able to overcome the drawbacks such as over specialization, data sparsity and inefficiency issue of most traditional recommender systems.