A Collaborative Filtering Recommendation Algorithm with Time Adjusting Based on Attribute Center of Gravity Model

Abstract

In the collaborative filtering recommendation technology, the similarity measurement part plays a vital role, and similarity measurement accuracy seriously affects the similarity measurement part and all the subsequent parts. However, there are many shortcomings in the similarity measurement part of traditional memory-based collaborative filtering recommendation technology. In order to solve the inaccuracy under special circumstances, this paper proposes an improved algorithm, a collaborative filtering recommendation algorithm with time adjusting based on attribute center of gravity model, through altering the process of similarity calculation. Simulation results show that the improved algorithm gains a higher recommendation accuracy, compared with the traditional algorithms.