A Study on Item-Based Collaborative Filtering Algorithm Using Semantic Similarity

In a recommendation system based on collaborative filtering(CF), in order to resolve efficiently some problems such as the scalability and sparsity problems the quality of recommendation system will tend to be decreased dramatically. A new CF algorithm based on semantic knowledge of items is presented. The algorithm takes synthetically into account the influence of item semantic and user rating to enhance the item-based CF. Experimental results indicate that the algorithm can achieve better prediction accuracy and provide better recommendation results than with the traditional CF algorithms.