Combining the Results of Several Neural Network Classifiers

  • Galina Rogova
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 219)


Neural networks and traditional classifiers work well for optical character recognition; however, it is advantageous to combine the results of several algorithms to improve classification accuracies. This paper presents a combination method based on the Dempster–Shafer theory of evidence, which uses statistical information about the relative classification strengths of several classifiers. Numerous experiments show the effectiveness of this approach. The method allows 15—30% reduction of misclassification error compared to the best individual classifier.


Classifier Neural network Character recognition The Dempster–Shafer theory of evidence Evidence 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Anderson, P.G., & Gaborski, R.S. (1993). The polynomial method augmented by supervised training for handprinted character recognition. Proceedings of the International Conference on Neural Networks and Genetic Algorithmspp. 417–422. Innsbruck, Austria: Springer-Verlag.Google Scholar
  2. 2.
    Barnett, J.A. (1981). Computational methods for mathematical theory of evidence. Proceedings of Seventh International Joint Conference on Artificial Intelligence pp. 868–875. Vancouver, BC.Google Scholar
  3. 3.
    Mandler, E.J., & Schurmann, J. (1988). Combining the classification results of independent classifiers based on the Dempster/Shafer theory of evidence. Pattern Recognition and Artificial Intelligence, X, 381–393.Google Scholar
  4. 4.
    Pawlicki, T.F. (1991). Personal communication.Google Scholar
  5. 5.
    Shafer, G. (1976). A mathematical theory of evidence. Princeton: MIT Press.MATHGoogle Scholar
  6. 6.
    Shustorovich, A. (in press). A subspace projection approach to feature extraction: the two-dimensional Gabor transform for character recognition. Neural Networks.Google Scholar
  7. 7.
    U.S. Census Bureau/NIST First OCR Systems Conference, May 27–28, 1992, Gaithersburg, MD.Google Scholar
  8. 8.
    Xu, L., Krzyzak, A., & Suen, C.Y. (1991). Associative switch for combining multiple classifiers. Proceedings of the International Joint Conference on Neural Networks pp. I-43–48. Seattle, WA: IEEE Press.Google Scholar
  9. 9.
    Xu, L., Krzyzak, A., & Suen, C.Y. (1992). Methods of combining multiple classifiers and their applications to handwriting recognition. IEEE Transactions on Systems, Man, and Cybernetics, 22(3), 418–435.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Galina Rogova

There are no affiliations available

Personalised recommendations