Likelihood-Based Robust Classification with Bayesian Networks

  • Alessandro Antonucci
  • Marco E. G. V. Cattaneo
  • Giorgio Corani
Part of the Communications in Computer and Information Science book series (CCIS, volume 299)

Abstract

Bayesian networks are commonly used for classification: a structural learning algorithm determines the network graph, while standard approaches estimate the model parameters from data. Yet, with few data the corresponding assessments can be unreliable. To gain robustness in this phase, we consider a likelihood-based learning approach, which takes all the model quantifications whose likelihood exceeds a given threshold. A new classification algorithm based on this approach is presented. Notably, this is a credal classifier, i.e., more than a single class can be returned in output. This is the case when the Bayesian networks consistent with the threshold constraint assign different class labels to a test instance. This is the first classifier of this kind for general topologies. Experiments show how this approach provide the desired robustness.

Keywords

Classification likelihood-based learning Bayesian networks credal networks imprecise probabilities credal classifiers 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Antonucci, A., Cattaneo, M., Corani, G.: Likelihood-based naive credal classifier. In: ISIPTA 2011, pp. 21–30. SIPTA (2011)Google Scholar
  2. 2.
    Antonucci, A., Cattaneo, M., Corani, G.: Likelihood-based robust classification with Bayesian networks. Technical Report IDSIA-01-12, IDSIA (2012)Google Scholar
  3. 3.
    Cattaneo, M.: Statistical Decisions Based Directly on the Likelihood Function. PhD thesis, ETH Zurich (2007)Google Scholar
  4. 4.
    Cattaneo, M.: Likelihood-based inference for probabilistic graphical models: Some preliminary results. In: PGM 2010, pp. 57–64. HIIT Publications (2010)Google Scholar
  5. 5.
    Corani, G., Antonucci, A., Zaffalon, M.: Bayesian networks with imprecise probabilities: Theory and application to classification. In: Data Mining: Foundations and Intelligent Paradigms, pp. 49–93. Elsevier (2010)Google Scholar
  6. 6.
    Cozman, F.G.: Credal networks. Artif. Intell. 120, 199–233 (2000)MathSciNetCrossRefMATHGoogle Scholar
  7. 7.
    Friedman, N., Geiger, D., Goldszmidt, M.: Bayesian networks classifiers. Mach. Learn. 29, 131–163 (1997)CrossRefMATHGoogle Scholar
  8. 8.
    Hudson, D.J.: Interval estimation from the likelihood function. J. R. Stat. Soc., Ser. B 33, 256–262 (1971)MathSciNetGoogle Scholar
  9. 9.
    Koller, D., Friedman, N.: Probabilistic Graphical Models. MIT Press (2009)Google Scholar
  10. 10.
    Moral, S.: Calculating uncertainty intervals from conditional convex sets of probabilities. In: UAI 1992, pp. 199–206. Morgan Kaufmann (1992)Google Scholar
  11. 11.
    Pawitan, Y.: In All Likelihood. Oxford University Press (2001)Google Scholar
  12. 12.
    Pearl, J.: Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann (1988)Google Scholar
  13. 13.
    Walley, P.: Statistical Reasoning with Imprecise Probabilities. Chapman and Hall (1991)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Alessandro Antonucci
    • 1
  • Marco E. G. V. Cattaneo
    • 2
  • Giorgio Corani
    • 1
  1. 1.Istituto Dalle Molle di Studi sull’Intelligenza Artificiale (IDSIA)Manno-LuganoSwitzerland
  2. 2.Department of StatisticsLMU MunichMünchenGermany

Personalised recommendations