An Analysis of Probabilistic Methods for Top-N Recommendation in Collaborative Filtering

  • Nicola Barbieri
  • Giuseppe Manco
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6911)


In this work we perform an analysis of probabilistic approaches to recommendation upon a different validation perspective, which focuses on accuracy metrics such as recall and precision of the recommendation list. Traditionally, state-of-art approches to recommendations consider the recommendation process from a “missing value prediction” perspective. This approach simplifies the model validation phase that is based on the minimization of standard error metrics such as RMSE. However, recent studies have pointed several limitations of this approach, showing that a lower RMSE does not necessarily imply improvements in terms of specific recommendations. We demonstrate that the underlying probabilistic framework offers several advantages over traditional methods, in terms of flexibility in the generation of the recommendation list and consequently in the accuracy of recommendation.


Recommender Systems Collaborative Filtering Probabilistic Topic Models Performance 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Agarwal, D., Chen, B.-C.: flda: matrix factorization through latent dirichlet allocation. In: WSDM, pp. 91–100 (2010)Google Scholar
  2. 2.
    Barbieri, N., Guarascio, M., Manco, G.: A probabilistic hierarchical approach for pattern discovery in collaborative filtering data. In: SMD (2011)Google Scholar
  3. 3.
    Blei, D.M., Ng, A.Y., Jordan, M.I.: Latent dirichlet allocation. The Journal of Machine Learning Research 3, 993–1022 (2003)zbMATHGoogle Scholar
  4. 4.
    Cremonesi, P., Koren, Y., Turrin, R.: Performance of recommender algorithms on top-n recommendation tasks. In: ACM RecSys, pp. 39–46 (2010)Google Scholar
  5. 5.
    Cremonesi, P., Turrin, R., Lentini, E., Matteucci, M.: An evaluation methodology for collaborative recommender systems. In: AXMEDIS, pp. 224–231 (2008)Google Scholar
  6. 6.
    Ge, M., Delgado-Battenfeld, C., Jannach, D.: Beyond accuracy: evaluating recommender systems by coverage and serendipity. In: ACM RecSys, pp. 257–260 (2010)Google Scholar
  7. 7.
    Goldberg, D., Nichols, D., Oki, B.M., Terry, D.: Using collaborative filtering to weave an information tapestry. Communications of the ACM 35(12), 61–70 (1992)CrossRefGoogle Scholar
  8. 8.
    Hofmann, T.: Collaborative filtering via gaussian probabilistic latent semantic analysis. In: SIGIR (2003)Google Scholar
  9. 9.
    Hofmann, T.: Latent semantic models for collaborative filtering. ACM Transactions on Information Systems (TOIS) 22(1), 89–115 (2004)CrossRefGoogle Scholar
  10. 10.
    Hofmann, T., Puzicha, J.: Latent class models for collaborative filtering. IJCAI, 688–693 (1999)Google Scholar
  11. 11.
    Jin, X., Zhou, Y., Mobasher, B.: A maximum entropy web recommendation system: combining collaborative and content features. In: KDD, pp. 612–617 (2005)Google Scholar
  12. 12.
    Koren, Y.: How useful is a lower rmse? (2007),
  13. 13.
    Marlin, B.: Modeling user rating profiles for collaborative filtering. In: NIPS (2003)Google Scholar
  14. 14.
    Marlin, B., Marlin, B.: Collaborative filtering: A machine learning perspective. Tech. rep., Department of Computer Science University of Toronto (2004)Google Scholar
  15. 15.
    McNee, S.M., Riedl, J., Konstan, J.A.: Being accurate is not enough: How accuracy metrics have hurt recommender systems. In: ACM SIGCHI Conference on Human Factors in Computing Systems, pp. 1097–1101 (2006)Google Scholar
  16. 16.
    Pazzani, M.J., Billsus, D.: Content-based recommendation systems. The Adaptive Web: Methods and Strategies of Web Personalization, 325–341 (2007)Google Scholar
  17. 17.
    Salakhutdinov, R., Mnih, A.: Bayesian probabilistic matrix factorization using markov chain monte carlo. In: ICML, pp. 880–887 (2008)Google Scholar
  18. 18.
    Salakhutdinov, R., Mnih, A.: Probabilistic matrix factorization. In: NIPS, pp. 1257–1264 (2008)Google Scholar
  19. 19.
    Sarwar, B., Karypis, G., Konstan, J., Reidl, J.: Item-based collaborative filtering recommendation algorithms. In: WWW, pp. 285–295 (2001)Google Scholar
  20. 20.
    Shan, H., Banerjee, A.: Generalized probabilistic matrix factorizations for collaborative filtering. In: ICDM (2010)Google Scholar
  21. 21.
    Stern, D.H., Herbrich, R., Graepel, T.: Matchbox: large scale online bayesian recommendations. In: WWW, pp. 111–120 (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Nicola Barbieri
    • 1
    • 2
  • Giuseppe Manco
    • 2
  1. 1.Department of Electronics, Informatics and SystemsUniversity of CalabriaRendeItaly
  2. 2.Institute for High Performance Computing and Networks (ICAR)Italian National Research CouncilRendeItaly

Personalised recommendations