Lei Sun
ABSTRACT
The rating prediction algorithm based on matrix factorization is one of the research hotspots. When the training data is increased, all data needs to be retrained to improve recommendation results with new changes. When the data is large, the cost of calculation and time will be higher. This paper proposed an incremental recommendation algorithm based on regularized matrix factorization (IncRMF), the algorithm deals with new users and items ratings incrementally, get the updated data of training and ratings prediction by optimizing the previous result, and significantly reduce the amount of calculation in the process of training. Theoretical analysis and experimental results show that this method can guarantee the accuracy of prediction results (the difference of prediction error between IncRMF and normal regularized matrix factorization method not more than 1.36%), while the amount of calculation is significantly reduced.
- Bokde, D., Girase, S., Mukhopadhyay, D. 2015. Matrix Factorization Model in Collaborative Filtering Algorithms: A Survey {J}. Procedia Computer Science, 49(1):136--146.Google Scholar
Cross Ref
- Koren, Y., Bell, R., Volinsky, C. 2009. Matrix Factorization Techniques for Recommender Systems{J}. Computer, 42(8):30--37. Google ScholarDigital Library
- Papagelis, M., Rousidis, I., Plexousakis, D. et al. 2005:5. Incremental Collaborative Filtering for Highly-Scalable Recommendation Algorithms{C}// International Symposium on Methodologies for Intelligent Systems. Springer Berlin Heidelberg. Google ScholarDigital Library
- Yang, X., Zhang, Z., Wang, K. 2012. Scalable collaborative filtering using incremental update and local link prediction{C}// ACM, 2371--2374. Google ScholarDigital Library
- Sarwar, B., Konstan, J., Riedl, J. 2002. Incremental singular value decomposition algorithms for highly scalable recommender systems{C}. In Fifth International Conference on Computer and Information Science, pages 27--28. Citeseer.Google Scholar
- Takács, G., Pilászy, I., Németh, B., and Tikk, D. 2009. Scalable collaborative filtering approaches for large recommender systems{J}. The Journal of Machine Learning Research, 10:623--656. Google ScholarDigital Library
- Rendle, S., Schmidt-Thieme, L. 2008. Online-updating regularized kernel matrix factorization models for large-scale recommender systems{C}// ACM Conference on Recommender Systems. ACM, 251--258. Google ScholarDigital Library
- Luo, X., Zhou, M. C., Leung, H. et al. 2016. An Incremental-and-Static-Combined Scheme for Matrix-Factorization-Based Collaborative Filtering{J}. IEEE Transactions on Automation Science & Engineering, 13(1):333--343.Google ScholarCross Ref
- Song, Q., Cheng, J., Lu, H. 2015. Incremental Matrix Factorization via Feature Space Re-learning for Recommender System{C}// The, ACM Conference.277--280. Google ScholarDigital Library
Index Terms
- IncRMF: An Incremental Recommendation Algorithm Based on Regularized Matrix Factorization
Recommendations
Adaptive matrix completion for the users and the items in tail
WWW '19: The World Wide Web ConferenceRecommender systems are widely used to recommend the most appealing items to users. These recommendations can be generated by applying collaborative filtering methods. The low-rank matrix completion method is the state-of-the-art collaborative filtering ...
Attributes coupling based matrix factorization for item recommendation
Recommender systems have attracted lots of attention since they alleviate the information overload problem for users. Matrix factorization is one of the most widely employed collaborative filtering techniques in the research of recommender systems due ...
Personalised rating prediction for new users using latent factor models
HT '11: Proceedings of the 22nd ACM conference on Hypertext and hypermediaIn recent years, personalised recommendations have gained importance in helping users deal with the abundance of information available online. Personalised recommendations are often based on rating predictions, and thus accurate rating prediction is ...
Comments