A User-Based Collaborative Filtering Recommendation Algorithm Based on Folksonomy Smoothing

Recommender systems suggest useful and interesting products to customers in order to increase customer satisfaction and online conversion rates. Collaborative filtering recommendation algorithm is the most usually applied recommender system for personalized. In view of the fact that collaborative filtering systems depend on neighbors as information sources, the recommendation quality of collaborative filtering relies on the neighbors selected. However, traditional collaborative filtering has some essential limitations in selecting neighbors. One of these chiefly problems is data sparsity. While the number of items is increase, the ratio of common rated items is decrease so calculating the computations of neighborhood become difficult. To alleviate the sparsity, a user-based collaborative filtering recommendation algorithm based on folksonomy smoothing is presented. The approach firstly fills the empty using folksonomy technology. And then produce the recommendations employing the user-based collaborative filtering algorithm.