Two Stage Reject Rule for ECOC Classification Systems

  • Paolo Simeone
  • Claudio Marrocco
  • Francesco Tortorella
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6713)


The original task of a multiclass classification problem can be decomposed using Error Correcting Output Coding in several two-class problems which can be solved with dichotomizers. A reject rule can be set on the classification system to improve the reliability of decision through an external threshold on the decoding outcomes before the decision is taken. If a loss-based decoding rule is used, more can be done to make such external scheme works better introducing a further reject stage in the system. This internal approach is meant to single out unreliable decisions for each classifier in order to proficiently exploit the properties of loss decoding techniques for ECOC as proved by experimental results on popular benchmarks.


ECOC reject option multiple classifiers systems 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Kong, E.B., Dietterich, T.G.: Error-correcting output coding corrects bias and variance. In: ICML, pp. 313–321 (1995)Google Scholar
  2. 2.
    Ghani, R.: Combining labeled and unlabeled data for text classification with a large number of categories. In: Cercone, N., Lin, T.Y., Wu, X. (eds.) ICDM, pp. 597–598. IEEE Computer Society Press, Los Alamitos (2001)Google Scholar
  3. 3.
    Zhou, J., Suen, C.Y.: Unconstrained numeral pair recognition using enhanced error correcting output coding: A holistic approach. In: ICDAR, pp. 484–488. IEEE Computer Society Press, Los Alamitos (2005)Google Scholar
  4. 4.
    Windeatt, T., Ardeshir, G.: Boosted ECOC ensembles for face recognition. In: Proceedings of the International Conference on Visual Information Engineering, pp. 165–168 (2003)Google Scholar
  5. 5.
    Kittler, J., Ghaderi, R., Windeatt, T., Matas, J.: Face verification via error correcting output codes. Image Vision Comput. 21(13-14), 1163–1169 (2003)CrossRefGoogle Scholar
  6. 6.
    Singh, S., Kodali, A., Choi, K., Krishna, R., Pattipati, Namburu, S.M., Sean, S.C., Prokhorov, D.V., Qiao, L.: Dynamic multiple fault diagnosis: Mathematical formulations and solution techniques. IEEE Transactions on Systems, Man, and Cybernetics, Part A 39(1), 160–176 (2009)CrossRefGoogle Scholar
  7. 7.
    Chow, C.: On optimum recognition error and reject tradeoff. IEEE Transactions on Information Theory 16(1), 41–46 (1970)CrossRefzbMATHGoogle Scholar
  8. 8.
    Simeone, P., Marrocco, C., Tortorella, F.: Exploiting system knowledge to improve ecoc reject rules. International Conference on Pattern Recognition 4340–4343 (2010)Google Scholar
  9. 9.
    Allwein, E.L., Schapire, R.E., Singer, Y.: Reducing multiclass to binary: A unifying approach for margin classifiers. Journal of Machine Learning Research 1, 113–141 (2000)zbMATHGoogle Scholar
  10. 10.
    Pietraszek, T.: On the use of ROC analysis for the optimization of abstaining classifiers. Machine Learning 68(2), 137–169 (2007)CrossRefGoogle Scholar
  11. 11.
    Asuncion, A., Newman, D.J.: UCI machine learning repository (2007)Google Scholar
  12. 12.
    Dietterich, T.G., Bakiri, G.: Solving multiclass learning problems via error-correcting output codes. Journal of Artificial Intelligence Research 2, 263–286 (1995)zbMATHGoogle Scholar
  13. 13.
    Vezhnevets, A., Vezhnevets, V.: Modest adaboost - teaching adaboost to generalize better. Graphicon 2005 (2005)Google Scholar
  14. 14.
    Joachims, T.: Making large-scale SVM learning practical. In: Schölkopf, B., Burges, C., Smola, A. (eds.) Advances in Kernel Methods - Support Vector Learning, ch. 11, MIT Press, Cambridge (1999)Google Scholar
  15. 15.
    Wilcoxon, F.: Individual comparisons by ranking methods. Biometrics Bulletin 1(6), 80–83 (1945)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Paolo Simeone
    • 1
  • Claudio Marrocco
    • 1
  • Francesco Tortorella
    • 1
  1. 1.DAEIMIUniversità degli Studi di CassinoCassinoItaly

Personalised recommendations