Abstract
Combining classifiers using Bayesian formalism deals with a high dimensional probability distribution composed of a class and the decisions of classifiers. Thus product approximation is needed for the probability distribution. Bayes error rate is upper bounded by the conditional entropy of the class and decisions, so the upper bound should be minimized for raising the class discrimination. By considering the dependency between class and decisions, dependency-based product approximation is proposed in this paper together with its related combination method. The proposed method is evaluated with the recognition of unconstrained handwritten numerals.
This research was financially supported by Hansung University in the year of 2004.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Xu, L., Krzyzak, A., Suen, C.Y.: Methods of Combining Multiple Classifiers and Their Applications to Handwriting Recognition. IEEE TSMC 22, 418–435 (1992)
Chow, C.K., Liu, C.N.: Approximating Discrete Probability Distributions with Dependence Trees. IEEE Trans. on Information Theory 14, 462–467 (1968)
Kang, H.J., Kim, K., Kim, J.H.: Optimal Approximation of Discrete Probability Distribution with kth-order Dependency and Its Applications to Combining Multiple Classifiers. PRL 18, 515–523 (1997)
Kang, H.J.: Combining multiple classifiers based on third-order dependency for handwritten numeral recognition. PRL 24, 3027–3036 (2003)
Wang, D.C.C., Wong, A.K.C.: Classification of Discrete Data with Feature Space Transform. IEEE TAC AC-24, 434–437 (1979)
Kang, H.J., Lee, S.W.: Combining Classifiers based on Minimization of a Bayes Error Rate. In: Proc. of the 5th ICDAR, pp. 398–401 (1999)
Huang, Y.S., Suen, C.Y.: A Method of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals. IEEE Trans. on PAMI 17, 90–94 (1995)
Suen, C.Y., Nadal, C., Legault, R., Mai, T.A., Lam, L.: Computer Recognition of Unconstrained Handwritten Numerals. Proc. of IEEE, 1162–1180 (1992)
Blake, C., Merz, C.: UCI repository of machine learning databases (1998)
Matsui, T., Tsutsumida, T., Srihari, S.N.: Combination of Stroke/Background Structure and Contour-direction Features in Handprinted Alphanumeric Recognition. In: Proc. of the 4th IWFHR, pp. 87–96 (1994)
Oh, I.S., Suen, C.Y.: Distance features for neural network-based recognition of handwritten characters. IJDAR 1, 73–88 (1998)
Hellman, M.E., Raviv, J.: Probability of error, equivocation, and the Chernoff bound. IEEE Trans. on Information Theory IT-16, 368–372 (1970)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kang, HJ. (2004). Combining Classifiers Using Dependency-Based Product Approximation with Bayes Error Rate. In: Roli, F., Kittler, J., Windeatt, T. (eds) Multiple Classifier Systems. MCS 2004. Lecture Notes in Computer Science, vol 3077. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-25966-4_11
Download citation
DOI: https://doi.org/10.1007/978-3-540-25966-4_11
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22144-9
Online ISBN: 978-3-540-25966-4
eBook Packages: Springer Book Archive