An optimal recommendation system for content personalization management [DTV applications]

This work develops a new optimal program recommendation system. An IFQC (interval feature-query combination) structure is proposed for user preference modeling. Based on explicit (instant query, manual modification) and implicit (viewing behavior) feedback information, the recommendation system can learn the user's preferences adaptively and intelligently.