Dynamic Recommender System: Using Cluster-Based Biases to Improve the Accuracy of the Predictions

Chapter
Part of the Studies in Computational Intelligence book series (SCI, volume 615)

Abstract

It is today accepted that matrix factorization models allow a high quality of rating prediction in recommender systems. However, a major drawback of matrix factorization is its static nature that results in a progressive declining of the accuracy of the predictions after each factorization. This is due to the fact that the new obtained ratings are not taken into account until a new factorization is computed, which can not be done very often because of the high cost of matrix factorization. In this paper, aiming at improving the accuracy of recommender systems, we propose a cluster-based matrix factorization technique that enables online integration of new ratings. Thus, we significantly enhance the obtained predictions between two matrix factorizations. We use finer-grained user biases by clustering similar items into groups, and allocating in these groups a bias to each user. The experiments we did on large datasets demonstrated the efficiency of our approach.

References

  1. Adomavicius, G., and J. Zhang. 2012. Impact of data characteristics on recommender systems performance. ACM Transactions on Management Information Systems 3(1): 3:1–3:17.Google Scholar
  2. Agarwal, D, B.-C. Chen, and P. Elango. 2010. Fast online learning through offline initialization for time-sensitive recommendation. In Proceedings of the 16th ACM SIGKDD international conference on knowledge discovery and data mining, KDD ’10, 703–712. New York: ACM.Google Scholar
  3. Amatriain, X., and J. Basilico. 2012. Netflix recommendations: Beyond the 5 stars, 2012. The Netflix Tech Blog.Google Scholar
  4. Bell, R., Y. Koren, and C. Volinsky. 2007. Modeling relationships at multiple scales to improve accuracy of large recommender systems. In Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, KDD ’07, 95–104. New York: ACM.Google Scholar
  5. Bell, R.M., J. Bennett, Y. Koren, and C. Volinsky. 2009. The million dollar programming prize. IEEE Spectrum 46: 28–33.CrossRefGoogle Scholar
  6. Bennett, J., S. Lanning, and N. Netflix. 2007. The netflix prize. In In KDD cup and workshop in conjunction with KDD.Google Scholar
  7. Bickson, D. 2011. Large scale matrix factorization—yahoo! kdd cup, 2011. Large Scale Machine Learning and Other Animals.Google Scholar
  8. Cao, B., D. Shen, J.-T. Sun, X. Wang, Q. Yang, and Z. Chen. 2007. Detect and track latent factors with online nonnegative matrix factorization. In Proceedings of the 20th international joint conference on artificial intelligence, 2689–2694. San Francisco: Morgan Kaufmann Publishers Inc.Google Scholar
  9. Chakraborty, P. 2009. A scalable collaborative filtering based recommender system using incremental clustering. In Advance Computing Conference, IACC 2009. IEEE International, 1526–1529.Google Scholar
  10. Dias, M.B., D. Locher, M. Li, W. El-Deredy, and P.J. Lisboa. 2008. The value of personalised recommender systems to e-business: a case study. In Proceedings of the 2008 ACM conference on recommender systems, RecSys ’08, 291–294. New York: ACM.Google Scholar
  11. Dror, G., N. Koenigstein, Y. Koren, and M. Weimer. 2011. The yahoo! music dataset and kdd-cup’11. In Proceedings of KDDCup 2011.Google Scholar
  12. Fleder, D.M., and K. Hosanagar. 2007. Recommender systems and their impact on sales diversity. In Proceedings of the 8th ACM conference on electronic commerce, EC ’07, 192–199. New York: ACM.Google Scholar
  13. Gemulla, R., E. Nijkamp, P.J. Haas, and Y. Sismanis. 2011. Large-scale matrix factorization with distributed stochastic gradient descent. In Proceedings of the 17th ACM SIGKDD international conference on knowledge discovery and data mining, KDD ’11, 69–77. New York: ACM.Google Scholar
  14. Herlocker, J.L., J.A. Konstan, L.G. Terveen, and J.T. Riedl. 2004. Evaluating collaborative filtering recommender systems. ACM Transactions Informatics Systems 22: 5–53.CrossRefGoogle Scholar
  15. Jambor, T., J. Wang, and N. Lathia. 2012. Using control theory for stable and efficient recommender systems. In Proceedings of the 21st international conference on World Wide Web, WWW ’12, 11–20. New York: ACM.Google Scholar
  16. Jannach, D., and K. Hegelich. 2009. A case study on the effectiveness of recommendations in the mobile internet. In RecSys, ed. L.D. Bergman, A. Tuzhilin, R.D. Burke, A. Felfernig, and L. Schmidt-Thieme, 205–208. ACM.Google Scholar
  17. Koenigstein, N., G. Dror, and Y. Koren. 2011. Yahoo! music recommendations: modeling music ratings with temporal dynamics and item taxonomy. In Proceedings of the 5th ACM conference on recommender systems, RecSys ’11, 165–172. New York: ACM.Google Scholar
  18. Kogan, J. 2007. Introduction to clustering large and high-dimensional data. New York: Cambridge University Press.MATHGoogle Scholar
  19. Kogan, J., C. Nicholas, and M. Teboulle. 2006. Grouping multidimensional data: recent advances in clustering. New York: Springer.CrossRefGoogle Scholar
  20. Koren, Y. 2007. How useful is a lower rmse? Netflix Prize Forum.Google Scholar
  21. Koren, Y. 2009. The bellkor solution to the netflix grand prize.Google Scholar
  22. Koren, Y. 2010. Collaborative filtering with temporal dynamics. Communications ACM 53(4): 89–97.CrossRefGoogle Scholar
  23. Koren, Y., R. Bell, and C. Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer 42: 30–37.CrossRefGoogle Scholar
  24. Linden, G., B. Smith, and J. York. 2003. Industry report: Amazon.com recommendations: Item-to-item collaborative filtering. IEEE Distributed Systems Online 4(1).Google Scholar
  25. Paterek, A. 2007. Improving regularized singular value decomposition for collaborative filtering. In Proceedings of KDD cup workshop at SIGKDD’07, 13th ACM international conference on knowledge discovery and data mining 39–42.Google Scholar
  26. Rendle, S., and L. Schmidt-Thieme. 2008. Online-updating regularized kernel matrix factorization models for large-scale recommender systems. In RecSys, ed. P. Pu, D.G. Bridge, B. Mobasher, and F. Ricci, 251–258. ACM.Google Scholar
  27. Sarwar, B., G. Karypis, J. Konstan, and J. Riedl. 2002. Incremental singular value decomposition algorithms for highly scalable recommender systems. In Proceedings of the 5th international conference in computers and information technology.Google Scholar
  28. Schafer, J.B., J. Konstan, and J. Riedi. 1999. Recommender systems in e-commerce. In Proceedings of the 1st ACM conference on electronic commerce, EC ’99, 158–166. New York: ACM.Google Scholar
  29. Su, X., and T.M. Khoshgoftaar. 2009. A survey of collaborative filtering techniques. Advances in Artificial Intelligence 2009: 4:2–4:2.Google Scholar
  30. Sun, Y., G. Liu, and K. Xu. 2010. A k-means-based projected clustering algorithm. In Proceedings of the 2010 third international joint conference on computational science and optimization—volume 01, CSO ’10, 466–470. Washington: IEEE Computer Society.Google Scholar
  31. Takács, G., I. Pilászy, B. Németh, and D. Tikk. 2008. Investigation of various matrix factorization methods for large recommender systems. In Proceedings of the 2nd KDD workshop on large-scale recommender systems and the netflix prize competition, NETFLIX ’08, 6:1–6:8. New York: ACM.Google Scholar
  32. Takács, G., I. Pilászy, B. Németh, and D. Tikk. 2009. Scalable collaborative filtering approaches for large recommender systems. Journal Machinery Learning Research 10: 623–656.Google Scholar
  33. TPC-Council. 2010. Tpc benchmark c, rev 5.11. Technical report, Transaction Processing Performance Council.Google Scholar
  34. Ziegler, C.-N., G. Lausen, and J.A. Konstan. 2008. On exploiting classification taxonomies in recommender systems. AI Communications 21(2–3): 97–125.Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  1. 1.Telecom ParisTechParisFrance
  2. 2.Sorbonne Universités, UPMC Univ ParisParisFrance

Personalised recommendations