An Improved Collaborative Filtering Algorithm to Improve Recommendation Accuracy and Protect User Privacy

The emergence of recommendation algorithm alleviates the problem of information overload to a certain extent, makes recommendations to users on the premise of protecting user privacy, and helps users to explore potential demands. An improved collaborative filtering algorithm is proposed for solving the problem that the relationship between recommendation time as well as accuracy cannot be weighed and how to protect user privacy. In this paper, the matrix dimension reduction technique of locally optimized singular value decomposition(SVD) and the K-means clustering technique are used to reduce the dimensions and cluster similar users in the user-item scoring matrix. The approximate difference matrix is used to represent the local structure of the scoring matrix to implement the local optimization. The locally optimized SVD technique can alleviate the problem of data sparsity and poor scalability in collaborative filtering by using fewer iterations. K-means clustering technique can greatly narrow the search range of neighbor sets and improve the recommendation speed. Therefore, collaborative filtering algorithms based on dimensionality reduction and clustering can generate recommendations in an accurate and real-time manner. The experimental results on the MovieLens dataset show that the algorithm can reduce the impact of data sparsity, protect user privacy and effectively improve the accuracy of recommendation.

[1]  Jia Zhang,et al.  An effective collaborative filtering algorithm based on user preference clustering , 2016, Applied Intelligence.

[2]  Mahdi Jalili,et al.  Recommender systems based on collaborative filtering and resource allocation , 2014, Social Network Analysis and Mining.

[3]  Pankoo Kim,et al.  Adaptive Collaborative Filtering Based on Scalable Clustering for Big Recommender Systems , 2016 .

[4]  Matevz Kunaver,et al.  Diversity in recommender systems - A survey , 2017, Knowl. Based Syst..

[5]  Kyoung-jae Kim,et al.  Recommender systems using cluster-indexing collaborative filtering and social data analytics , 2017, Int. J. Prod. Res..

[6]  Zhang Yi,et al.  A Novel Deep Learning-Based Collaborative Filtering Model for Recommendation System , 2019, IEEE Transactions on Cybernetics.

[7]  Ruck Thawonmas,et al.  Bounded-SVD: A Matrix Factorization Method with Bound Constraints for Recommender Systems , 2015, 2015 International Conference on Emerging Information Technology and Engineering Solutions.

[8]  Chang-Tsun Li,et al.  Matrix Factorization With Rating Completion: An Enhanced SVD Model for Collaborative Filtering Recommender Systems , 2017, IEEE Access.

[9]  Rajiv Pandey,et al.  Elective Recommendation Support through K-Means Clustering Using R-Tool , 2015, 2015 International Conference on Computational Intelligence and Communication Networks (CICN).

[10]  F. Maxwell Harper,et al.  The MovieLens Datasets: History and Context , 2016, TIIS.

[11]  Mehrbakhsh Nilashi,et al.  Collaborative filtering recommender systems , 2013 .

[12]  Kourosh Kiani,et al.  User based Collaborative Filtering using fuzzy C-means , 2016 .

[13]  Diego Fernández,et al.  Comparison of collaborative filtering algorithms , 2011, ACM Trans. Web.