An Optimization Method for Recommendation System Based on User Implicit Behavior

Collaborative filtering is the most worldwide and personalized video recommendation technology. As collaborative filtering recommendation system is often faced with the problem of matrix sparse on user rating. Via the introduction of the concept of collaborative filtering and the analysis of user behaviors and solution to the problem of sparse existing recommendation systems, this paper puts forward with an optimization algorithm, combining with implicit user behavior and verify the effectiveness of the optimization algorithm through the experiment.