Multi-criteria collaborative recommender

Abstract

Collaborative filtering algorithm (CF) is a personalized recommendation algorithm that is the most widely used in e-commerce. In the process of collaborative filtering recommendation, the most used information is the item rating. Item attribute information and other criteria of item evaluation are rarely used. In this paper, a collaborative filtering algorithm based on collaboration between rating item and item information is proposed. The objective is to consider, not only item rating information when we calculate similarity, but also the integration of the background of the item and the time-weight as criteria of the item assessment. In so doing, the calculation of the similarity between items, forming the neighborhood of item, performs the recommendation. This proposed CF algorithm is showing to avoid the problem of sparsity, and also reduces the influence of the former evaluation of the item.