Abstract
The diversity of application domains of pattern recognition makes it difficult to find a highly reliable classification algorithm for sufficiently interesting tasks. In this paper we propose a new combining method, which harness the local confidence of each classifier in the combining process. Our method is at the confluence of two main streams of combining multiple classifiers: classifier fusion and classifier selection. This method learns the local confidence of each classifier using training data and if an unknown data is given, the learned knowledge is used to evaluate the outputs of individual classifiers. An empirical evaluation using five real data sets has shown that this method achieves a promising performance and outperforms the best single classifiers and other known combining methods we tried.
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.
References
Gader, D., Hepp, D., Forester, B., Peurach, T., Mitchell, T.: Pipelined systems for recognition of handwritten digits in USPS ZIP codes. In: Proc. of U.S. Postal Service Advanced Technology Conference, pp. 539–548 (1990)
Kimura, F., Shridhar, M.: Handwritten Numeral Recognition Based on Multiple Algorithms. Pattern Recognition 24(10), 969–983 (1991)
Matsui, T., Noumi, T., Yamashita, I., Wakahara, T., Yoshimuro, M.: State of the Art of Handwritten Numeral Recognition in japan-The results of the First IPTP Character Recognition Competition. In: Proc. of the 2nd ICDAR, pp. 391–396 (1993)
Noumi, T., et al.: Result of Second IPTP Character Recognition Competition and Studies on Multi-Expert Handwritten Numeral Recognition. In: Proc. of 4th IWFHR, pp. 338–346 (1994)
Xu, L., Krzyzak, A., Suen, Y.: Method of Combining Multiple Classifiers and Their Application to Handwritten Numeral Recognition. IEEE Trans. on Systems, Man and Cybernetics 22(3), 418–435 (1992)
Tumer, K., Ghosh, J.: Linear and order statistics combiners for pattern classification. In: Sharkey, A.J.C. (ed.) Combining Artificial Neural Nets, pp. 127–161 (1999)
Tumer, K., Gosh, J.: Error correlation and error reduction in ensemble classifiers, Tech. Report, Dept. of ECE, University of Texas, July 11 (1996)
Jordan, M.I., Jacobs, R.A.: Hierarchical Mixtures of Experts and the EM Algorithm. Neural Computation 6, 181–214 (1994)
Jordan, M.I., Jacobs, R.A.: Modular and Hierarchical Learning Systems. In: Arbib, M.A. (ed.) The Handbook of Brain Theory and Neural Networks (1995)
Avnimelech, R., Intrator, N.: Boosted Mixture of Experts: An Ensemble Learning Scheme. Neural Computation 11(2), 483–497 (1999)
Tang, B., Heywood, M.I., Shepherd, M.: Input partitioning to mixture of experts. In: International Joint Conference on Neural Networks, pp. 227–232 (2002)
Stern, H.: Improving on the mixture of experts algorithm, CSCI 6508: Fundamentals of Computational Neuroscience project, Dalhousie University, Halifax, NS, Canada (2003)
Woods, K.: Combination of Multiple Classifiers Using Local Accuracy Estimates. IEEE Trans. on Pattern Analysis and Machine Intelligence 19(4), 405–410
Mitiche, M., Thomas, D., Nagy, G.: Classifier Combination for Handprinted Digit Recognition. In: Proc. of 2nd Int’l Conf. on DAR, pp. 163–166 (1993)
Tsymbal, A.: Decision Committee Learning with Dynamic Integration of Classifier. LNCS. Springer, Heidelberg (2000)
Mitchell, T.: Machine Learning. Mcgraw Hill, New York (1997)
Rumelhard, E., Hinton, E., Williams, J.: Learning internal representations by error propagation. In: Parallel Distributed Processing: Explorations in the Microstructures of Cognition. MIT Press, Cambridge (1996)
Blake, C., Keogh, E., Merz, J.: UCI repository of Machine Learning databases (1999), http://www.ics.uci.edu/~mlearn/MLRepository.html
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2005 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kim, E., Ko, J. (2005). Dynamic Classifier Integration Method. In: Oza, N.C., Polikar, R., Kittler, J., Roli, F. (eds) Multiple Classifier Systems. MCS 2005. Lecture Notes in Computer Science, vol 3541. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11494683_10
Download citation
DOI: https://doi.org/10.1007/11494683_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-26306-7
Online ISBN: 978-3-540-31578-0
eBook Packages: Computer ScienceComputer Science (R0)