The smart recommender is a component that provides personalized
recommendations of resources to the user. In the IPTV context the
smart recommender is used to identify the most relevant TV shows and
videos for a user based on his interests, watching behavior, social
networks, and the feedback provided to the system.

The smart recommender can be divided in two parts. The semantic
store and the recommendation engine. The semantic store holds
information about resources and is based on the Resource Description
Framework (RDF) while the recommendation engine provides several
algorithms for the classification of contents and users, and for the
similarity computation between resources and for collaborative
filtering. Machine Learning techniques are applied so that the smart
recommender adapts to the user needs over time, increasing the
quality of recommendations. An integrated feedback collection
function allows the user to exert influence on the operating mode of
the recommender.