Classification Trees Based on Belief Functions

  • Nicolas Sutton-Charani
  • Sébastien Destercke
  • Thierry Denœux
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 164)

Abstract

Decision tree classifiers are popular classification methods. In this paper, we extend to multi-class problems a decision tree method based on belief functions previously described for two-class problems only. We propose three possible extensions: combining multiple two-class trees together and directly extending the estimation of belief functions within the tree to the multi-class setting. We provide experiment results and compare them to usual decision trees.

Keywords

Decision Tree Information Gain Belief Function Approximate Reasoning Conditional Belief 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Abellan, J., Moral, S.: Upper entropy of credal sets. Applications to credal classification. International Journal of Approximate Reasoning 39(2-3), 235–255 (2005)MathSciNetMATHCrossRefGoogle Scholar
  2. 2.
    Breiman, Friedman, Olshen, Stone: Classification And Regression Trees. Wadsworth, Belmont (1984)MATHGoogle Scholar
  3. 3.
    Ciampi, A.: Growing a tree classifier with imprecise data. Pattern Recognition Letters 21(9), 787–803 (2000)CrossRefGoogle Scholar
  4. 4.
    de Campos, L., Huete, J., Moral, S.: Probability intervals: a tool for uncertain reasoning. Int. J. Uncertainty Fuzziness Knowledge-Based Syst. 1, 167–196 (1994)Google Scholar
  5. 5.
    Dempster, A.P.: New methods for reasoning towards posterior distributions based on sample data. Annals of Mathematical Statistics 37, 355–374 (1966)MathSciNetMATHCrossRefGoogle Scholar
  6. 6.
    Denoeux, T.: Constructing belief functions from sample data using multinomial confidence regions. International Journal of Approximate Reasoning 42(3), 228–252 (2006)MathSciNetMATHCrossRefGoogle Scholar
  7. 7.
    Denoeux, T.: Maximum likelihood estimation from uncertain data in the belief function framework. IEEE Trans. on Know. and Data Eng. (2011) (to appear), doi:10.1109/TKDE.2011.201Google Scholar
  8. 8.
    Denœux, T., Bjanger, M.S.: Induction of decision trees from partially classified data using belief functions. In: 2000 IEEE International Conference on Systems, Man, and Cybernetics, vol. 4, pp. 2923–2928. IEEE (2000)Google Scholar
  9. 9.
    Elouedi, Z., Mellouli, K., Smets, P.: Belief decision trees: theoretical foundations. International Journal of Approximate Reasoning 28(2-3), 91–124 (2001)MathSciNetMATHCrossRefGoogle Scholar
  10. 10.
    Klir, G.J.: Uncertainty and information: foundations of generalized information theory. Wiley-IEEE Press (2006)Google Scholar
  11. 11.
    Quinlan, J.: Induction of decision trees. Machine Learning 1, 81–106 (1986)Google Scholar
  12. 12.
    Quost, B., Denœux, T.: Pairwise Classifier Combination using Belief Functions. Pattern Recognition Letters 28, 644–653 (2006)CrossRefGoogle Scholar
  13. 13.
    Shafer, G.: A Mathematical Theory of Evidence. Princeton University Press, Princeton (1976)MATHGoogle Scholar
  14. 14.
    Utkin, L.V.: Extensions of belief functions and possibility distributions by using the imprecise dirichlet model. Fuzzy Sets and Systems 154(3), 413–431 (2005)MathSciNetMATHCrossRefGoogle Scholar
  15. 15.
    Vannoorenbergue, P., Denœux, T.: Handling uncertain labels in multiclass problems using belief decision trees. In: IPMU 2002, Annecy, France, vol. 3, pp. 1919–1926 (2002)Google Scholar
  16. 16.
    Walley, P.: Inferences from multinomial data: Learning about a bag of marbles. Journal of the Royal Statistical Society, Series B, 3–57 (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Nicolas Sutton-Charani
    • 1
  • Sébastien Destercke
    • 1
  • Thierry Denœux
    • 1
  1. 1.UMR CNRS 7253 HeudiasycUniversité Technologique de CompiègneCompiègne cedexFrance

Personalised recommendations