Fast ALS-Based Tensor Factorization for Context-Aware Recommendation from Implicit Feedback

  • Balázs Hidasi
  • Domonkos Tikk
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7524)

Abstract

Albeit the implicit feedback based recommendation problem—when only the user history is available but there are no ratings—is the most typical setting in real-world applications, it is much less researched than the explicit feedback case. State-of-the-art algorithms that are efficient on the explicit case cannot be straightforwardly transformed to the implicit case if scalability should be maintained. There are few implicit feedback benchmark datasets, therefore new ideas are usually experimented on explicit benchmarks. In this paper, we propose a generic context-aware implicit feedback recommender algorithm, coined iTALS. iTALS applies a fast, ALS-based tensor factorization learning method that scales linearly with the number of non-zero elements in the tensor. The method also allows us to incorporate various contextual information into the model while maintaining its computational efficiency. We present two context-aware implementation variants of iTALS. The first incorporates seasonality and enables to distinguish user behavior in different time intervals. The other views the user history as sequential information and has the ability to recognize usage pattern typical to certain group of items, e.g. to automatically tell apart product types that are typically purchased repetitively or once. Experiments performed on five implicit datasets (LastFM 1K, Grocery, VoD, and “implicitized” Netflix and MovieLens 10M) show that by integrating context-aware information with our factorization framework into the state-of-the-art implicit recommender algorithm the recommendation quality improves significantly.

Keywords

recommender systems tensor factorization context awareness implicit feedback 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Pilászy, I., Tikk, D.: Recommending new movies: Even a few ratings are more valuable than metadata. In: Recsys 2009: ACM Conf. on Recommender Systems, New York, NY, USA, pp. 93–100 (2009)Google Scholar
  2. 2.
    Bennett, J., Lanning, S.: The Netflix Prize. In: KDD Cup Workshop at SIGKDD 2007, San Jose, California, USA, pp. 3–6 (2007)Google Scholar
  3. 3.
    Bell, R.M., Koren, Y.: Scalable collaborative filtering with jointly derived neighborhood interpolation weights. In: ICDM 2007: IEEE Int. Conf. on Data Mining, Omaha, NE, USA, pp. 43–52 (2007)Google Scholar
  4. 4.
    Takács, G., Pilászy, I., Németh, B., Tikk, D.: Major components of the Gravity recommendation system. SIGKDD Explor. Newsl. 9, 80–83 (2007)CrossRefGoogle Scholar
  5. 5.
    Pilászy, I., Zibriczky, D., Tikk, D.: Fast ALS-based matrix factorization for explicit and implicit feedback datasets. In: Recsys 2010: ACM Conf. on Recommender Systems, Barcelona, Spain, pp. 71–78 (2010)Google Scholar
  6. 6.
    Takács, G., Pilászy, I., Tikk, D.: Applications of the conjugate gradient method for implicit feedback collaborative filtering. In: RecSys 2011: ACM Conf. on Recommender Systems, Chicago, IL, USA, pp. 297–300 (2011)Google Scholar
  7. 7.
    Koren, Y.: Factorization meets the neighborhood: a multifaceted collaborative filtering model. In: SIGKDD 2008: ACM Int. Conf. on Knowledge Discovery and Data Mining, Las Vegas, Nevada, USA, pp. 426–434 (2008)Google Scholar
  8. 8.
    Salakhutdinov, R., Mnih, A.: Probabilistic matrix factorization. In: Platt, J.C., Koller, D., Singer, Y., Roweis, S. (eds.) Advances in Neural Information Processing Systems 20. MIT Press, Cambridge (2008)Google Scholar
  9. 9.
    Ricci, F., Rokach, L., Shapira, B.: Introduction to recommender systems handbook. In: Recommender Systems Handbook, pp. 1–35. Springer, US (2011)CrossRefGoogle Scholar
  10. 10.
    Adomavicius, G., Tuzhilin, A.: Context-aware recommender systems. In: Recsys 2008: ACM Conf. on Recommender Systems, Lausanne, Switzerland, pp. 335–336 (2008)Google Scholar
  11. 11.
    Karatzoglou, A., Amatriain, X., Baltrunas, L., Oliver, N.: Multiverse recommendation: N-dimensional tensor factorization for context-aware collaborative filtering. In: Recsys 2010: ACM Conf. on Recommender Systems, Barcelona, Spain, pp. 79–86 (2010)Google Scholar
  12. 12.
    Adomavicius, G., Sankaranarayanan, R., Sen, S., Tuzhilin, A.: Incorporating contextual information in recommender systems using a multidimensional approach. ACM Trans. Inf. Syst. 23(1), 103–145 (2005)CrossRefGoogle Scholar
  13. 13.
    Adomavicius, G., Ricci, F.: Workshop on context-aware recommender systems (CARS-2009). In: Recsys 2009: ACM Conf. on Recommender Systems, New York, NY, USA, pp. 423–424 (2009)Google Scholar
  14. 14.
    Said, A., Berkovsky, S., De Luca, E.W.: Putting things in context: Challenge on context-aware movie recommendation. In: CAMRa 2010: Workshop on Context-Aware Movie Recommendation, Barcelona, Spain, pp. 2–6 (2010)Google Scholar
  15. 15.
    Bogers, T.: Movie recommendation using random walks over the contextual graph. In: CARS 2010: 2nd Workshop on Context-Aware Recommender Systems, Barcelona, Spain, pp. 1–5 (2010)Google Scholar
  16. 16.
    Baltrunas, L., Amatriain, X.: Towards time-dependant recommendation based on implicit feedback. In: CARS 2009: Workshop on Context-aware Recommender Systems, New York, NY, USA, pp. 1–5 (2009)Google Scholar
  17. 17.
    Bader, R., Neufeld, E., Woerndl, W., Prinz, V.: Context-aware POI recommendations in an automotive scenario using multi-criteria decision making methods. In: CaRR 2011: Workshop on Context-awareness in Retrieval and Recommendation, Palo Alto, CA, USA, pp. 23–30 (2011)Google Scholar
  18. 18.
    He, Q., Pei, J., Kifer, D., Mitra, P., Giles, L.: Context-aware citation recommendation. In: WWW 2010: Int. Conf. on World Wide Web, Raleigh, NC, USA, pp. 421–430 (2010)Google Scholar
  19. 19.
    Panniello, U., Tuzhilin, A., Gorgoglione, M., Palmisano, C., Pedone, A.: Experimental comparison of pre- vs. post-filtering approaches in context-aware recommender systems. In: Recsys 2009: ACM Conf. on Recommender Systems, New York, NY, USA, pp. 265–268 (2009)Google Scholar
  20. 20.
    Lathauwer, L.D., Moor, B.D., Vandewalle, J.: A multilinear singular value decomposition. SIAM J. Matrix Anal. Appl. 21(4), 1253–1278 (2000)MathSciNetMATHCrossRefGoogle Scholar
  21. 21.
    Rendle, S., Gantner, Z., Freudenthaler, C., Schmidt-Thieme, L.: Fast context-aware recommendations with factorization machines. In: SIGIR 2011: ACM Int. Conf. on Research and Development in Information, Beijing, China, pp. 635–644 (2011)Google Scholar
  22. 22.
    Hu, Y., Koren, Y., Volinsky, C.: Collaborative filtering for implicit feedback datasets. In: ICDM 2008: IEEE Int. Conf. on Data Mining, Pisa, Italy, pp. 263–272 (2008)Google Scholar
  23. 23.
    Liu, N.N., Cao, B., Zhao, M., Yang, Q.: Adapting neighborhood and matrix factorization models for context aware recommendation. In: CAMRa 2010: Workshop on Context-Aware Movie Recommendation, Barcelona, Spain, pp. 7–13 (2010)Google Scholar
  24. 24.
    Agrawal, R., Imieliński, T., Swami, A.: Mining association rules between sets of items in large databases. In: SIGMOD 1993: ACM SIGMOD Int. Conf. on Management of Data, Washington DC, USA, pp. 207–216 (1993)Google Scholar
  25. 25.
    Davidson, J., Liebald, B., Liu, J., et al.: The YouTube video recommendation system. In: Recsys 2010: ACM Conf. on Recommender Systems, Barcelona, Spain, pp. 293–296 (2010)Google Scholar
  26. 26.
    Celma, O.: Music Recommendation and Discovery in the Long Tail. Springer (2010)Google Scholar
  27. 27.
    GroupLens Research: Movielens data sets (2006), http://www.grouplens.org/node/73

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Balázs Hidasi
    • 1
    • 2
  • Domonkos Tikk
    • 1
  1. 1.Gravity R&D Ltd.Hungary
  2. 2.Budapest University of Technology and EconomicsHungary

Personalised recommendations