Abstract
When the sample size is small, the optimistically biased outputs produced by expert classifiers create serious problems for the combiner rule designer. To overcome these problems, we derive analytical expressions for bias reduction for situations when the standard Gaussian density-based quadratic classifiers serve as experts and the decisions of the base experts are aggregated by the behavior-space-knowledge (BKS) method. These reduction terms diminish the experts’ overconfidence and improve the multiple classification system’s generalization ability. The bias-reduction approach is compared with the standard BKS, majority voting and stacked generalization fusion rules on two real-life datasets for which the different base expert aggregates comprise the multiple classification system.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
J. Kittler, F. Roli (eds). Multiple Classifier Systems. Lecture Notes in Computer Science, Springer Vol. 1857 (2000), vol. 2096 (2001), Vol. 2364 (2002).
J. Kittler. Combining classifiers: a theoretical framework. Pattern Analysis and Applications, vol. 1, pp.18–27, 1998
J. Kittler. A framework for classifier fusion: is still needed? Advances in Pattern Recognition, Lecture Notes in Computer Science, Springer, vol. 1876, pp. 45–56, 2000.
J. Gosh. Multi-classifier systems: back to the future. Multiple Classifier Systems. Lecture Notes in Computer Science, Springer, vol. 2364, pp. 1–15, 2002.
T.K. Ho. Data complexity analysis for classifier combination. Multiple Classifier Systems. Lecture Notes in Computer Science, Springer, vol. 2096, pp. 53–67, 2001.
S. Raudys. Multiple classifier systems in the context of feature extraction and selection. Multiple Classification Systems, Lecture Notes in Computer Science, Springer, vol. 2364, pp. 27–41, 2002.
J. Kittler, M. Hatef, R.P.W. Duin, J. Matas. On combining classifiers, IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 20, pp. 226–239, 1998.
L.I. Kuncheva, J.C. Bezdek, R.P.W. Duin. Decision templates for multiple classifier fusion: and experimental comparison. Pattern Recognition, vol. 34, pp. 299–314, 2001.
L. Breiman. Bagging predictors. Machine Learning Journal, vol. 24, pp. 123–140, 1996.
L. Breiman. Arcing classifiers. Annals of Statistics, vol. 26, pp. 801–849, 1998.
Y. Freund, R.E. Schapire. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and Systems Sci., vol. 55, pp. 119–139, 1997.
M. Jordan and R. Jakobs. Hierarchical mixture of experts and the EM algorithm. Neural Computation, vol. 6, pp. 181–214, 1994.
D. Wolpert. Stacked generalization. Neural Networks, vol. 5, pp. 241–260, 1992.
F. Roli, G. Fumera. Analysis of linear and order statistics combiners for fusion of imbalanced classifiers. Multiple Classification Systems, Lecture Notes in ComputerScience, Springer, vol. 2364, pp. 252–261, 2002.
F. Roli, S. Raudys, G.L. Marcialis. An experimental comparison of fixed and trained rules for crisp classifiers outputs, Multiple Classification Systems, Lecture Notes in Computer Science, Springer, vol. 2364, pp. 232–241, 2002.
C. Güler, B. Sankur, Y. Kahya, M. Skurichina, S. Raudys. Classification of respiratory sound patterns by means of cooperative neural networks. In: G. Ramponi, G. L. Sicuranza, S. Carrato, S. Marsi (eds), Proc. of 8th European Signal Processing Conference (isbn 88-86179-83-9). Edizioni Lint, Trieste, 1996.
A. Janeliunas, S. Raudys. Reduction of boasting bias’ of linear expert. Multiple Classification Systems, Springer, Lecture Notes in Comp. Science, vol. 2364, pp. 242–251,2002.
S. Raudys. Experts’ boasting in trainable fusion rules. IEEE Transactions on Pattern Analysis and Machine Intelligence (in press, 2003).
S. Raudys. Statistical and Neural Classifiers: An integrated approach to design. Springer, London, p. 312, 2001.
P.A. Lachenbruch, M. Goldstein. Discriminant analysis. Biometrics, vol. 5, pp. 9–85, 1979.
Y.S. Huang, C.Y. Suen. A method of combining multiple experts for the recognition of unconstrained handwritten numerals. IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 17, pp. 90–93, 1998.
L.I. Kuncheva, C.J. Whitaker. Feature subsets for classifier combination: An enumerative experiment. Multiple Classifier Systems. Lecture Notes in Computer Science, Springer, vol. 2096, pp. 228–237, 2001.
K. Fukunaga. Introduction to Statistical Pattern Recognition. 2nd ed. Academic Press, New York, 1990.
R.O. Duda, P.E. Hart, D.G. Stork. Pattern Classification. 2nd ed. Wiley, NY, 2000.
S. Raudys, (1972). On the amount of a priori information in designing the classification algorithm. Proc. Acad. of Sciences of the USSR, 168–174, (in Russian).
S. Raudys and D. Young. A review of the former Soviet Union literature on small training sample problems in statistical discriminant analysis. Journal of Multivariate Analysis. (in press 2003).
A.E. Nikulin, B. Dolenko, T. Bezabeh, R.L. Somorjai. Near-optimal region selection for feature space reduction: novel preprocessing methods for classifying MR spectra. NMR Biomed. 11, 209–216, 1998.
C.L. Lean, R.L Somorjai, I.C.P. Smith, P. Russell, C.E. Mountford. Accurate diagnosis and prognosis of human cancers by proton MRS and a three stage classification strategy. Annual Reports on NMR Spectroscopy, vol. 48, 71–111, 2002.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Raudys, Š., Somorjai, R., Baumgartner, R. (2003). Reducing the Overconfidence of Base Classifiers when Combining Their Decisions. In: Windeatt, T., Roli, F. (eds) Multiple Classifier Systems. MCS 2003. Lecture Notes in Computer Science, vol 2709. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44938-8_7
Download citation
DOI: https://doi.org/10.1007/3-540-44938-8_7
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40369-2
Online ISBN: 978-3-540-44938-6
eBook Packages: Springer Book Archive