Parameter Estimation Method Based on Simulated Recommendation Process for Collaborative Filtering

In many E-commerce sites, recommender systems, which provide personalized recommendation from among a large number of items, are recently introduced. Collaborative filtering (CF) is one of the most successful algorithms which provide recommendations using ratings of users on items. There are two approaches such as user-based CF and item-based CF. Additionally a unifying method for user-based and item-based CF was proposed to improve the recommendation accuracy. The unifying approach uses a constant value as a weight parameter to unify both algorithms. However, because the optimal weight for unifying is actually different by the situation, the algorithm should estimate an appropriate weight dynamically, and should use it. In this research, we propose an approach for estimation of the appropriate weight based on collected ratings. Moreover, we discussed the effectiveness based on both multi-agent simulation and MovieLens dataset.