Discovering Rules for Dynamic Configuration of Multi-classifier Systems
This paper addresses the problem of dynamic configuration of multiclassifier systems. For this purpose, the performance of combination methods for abstract-level classifiers is predicted, under different working conditions, and sets of rules are discovered and used for dynamic configuration of multiclassifier systems. The experimental tests have been carried out in the field of hand-written numeral recognition. The result demonstrates the validity of the proposed approach.
- 2.T. Matsui, I. Yamashita, T. Wakahara, “Results of Second IPTP Character Recognition Competition and Studies on Multi-Expert Handwritten Numeral Recognition”, Proc. of IWFHR-4, 1994, pp.338–346.Google Scholar
- 3.Huang, C.Y. Suen, “An Optimal Method of Combining Multiple Classifiers for Unconstrained Handwritten Numeral Recognition”, Proc. of IWFHR-3, Buffalo, NY, 1993, pp. 11–20.Google Scholar
- 4.Y. Lu, F. Yamaoka, “Integration of Handwritten Digit Recognition results using Evidential Reasoning”, Proc. of IWFHR-4, 1994, pp. 456–463.Google Scholar
- 6.G. Dimauro, S. Impedovo, G. Pirlo, “Multiple Experts:A New Methodology for the Evaluation of the Combination Processes”, IWFHR-5, Colchester,Uk, 1996, pp.131–136.Google Scholar
- 7.V. Di Lecce, G. Dimauro, A. Guerriero, S. Impedovo, G. Pirlo, A. Salzo, “Classifier Combination: the role of a-priori knowledge”, Proc. of IWFHR-7, Sept. 2000, Amsterdam, The Netherlands, pp. 143–152.Google Scholar
- 8.S. Impedovo and A. Salzo, “A new methodology for expert combination in multi-expert system designing”. In Lecture Notes in Computer Science, (vol. 1857), J. Kittler and F. Roli Eds., MCS2000, Cagliari, Italy, pp.230–239.Google Scholar