A New Algorithm for Performing Ratings-Based Collaborative Filtering

Collaborative filtering is the most successful recommender system technology to date. It has been shown to produce high quality recommendations, but the performance degrades with the number of customers and products. In this paper, according to the feature of the rating data, we present a new similarity function Hsim(), and a signature table-based Algorithm for performing collaborative filtering. This method partitions the original data into sets of signature, then establishes a signature table to avoid a sequential scan. Our preliminary experiments based on a number of real data sets show that the new method can both improve the scalability and quality of collaborative filtering. Because the new method applies data clustering algorithms to rating data, predictions can be computed independently within one or a few partitions. Ideally, partition will improve the quality of collaborative filtering predictions. We'll continue to study how to further improve the quality of predictions in the future research.