SVD-based incremental approaches for recommender systems

• We propose an incremental algorithm called Incremental ApproSVD.
• It can predict unknown ratings when new items are entering dynamically.
• It is a suboptimal approximation with lower running time.
• We give the upper bound of error generated by Incremental ApproSVD.
• Experiments show the advantages of our algorithm on two real datasets.

Due to the serious information overload problem on the Internet, recommender systems have emerged as an important tool for recommending more useful information to users by providing personalized services for individual users. However, in the “big data” era, recommender systems face significant challenges, such as how to process massive data efficiently and accurately. In this paper we propose an incremental algorithm based on singular value decomposition (SVD) with good scalability, which combines the Incremental SVD algorithm with the Approximating the Singular Value Decomposition (ApproSVD) algorithm, called the Incremental ApproSVD. Furthermore, strict error analysis demonstrates the effectiveness of the performance of our Incremental ApproSVD algorithm. We then present an empirical study to compare the prediction accuracy and running time between our Incremental ApproSVD algorithm and the Incremental SVD algorithm on the MovieLens dataset and Flixster dataset. The experimental results demonstrate that our proposed method outperforms its counterparts.