A Relational Compound Collaborative Filtering Recommendation System

This paper presents a relational compound collaborative filtering (RCCF) recommendation system architecture, which integrated the behaviors associated mechanism and recommendation system to calculate the area associated values and the corresponding region of the recommended items rating, resulting in top-N list of the recommended item values. This system can avoid interested items in the MU¡¦s opposite direction and give higher weights value to the points of interest (POIs) in the predicted moving region in order to provide correct and real-time information and promote user satisfaction. We will use simulations to further verify the recommended accuracy and efficiency of the proposed RCCF recommendation system, but also look forward to apply on related mobile services.