This Article 
 Bibliographic References 
 Add to: 
Combination of Multiple Classifiers Using Local Accuracy Estimates
April 1997 (vol. 19 no. 4)
pp. 405-410

Abstract—This paper presents a method for combining classifiers that uses estimates of each individual classifier's local accuracy in small regions of feature space surrounding an unknown test sample. An empirical evaluation using five real data sets confirms the validity of our approach compared to some other Combination of Multiple Classifiers algorithms. We also suggest a methodology for determining the best mix of individual classifiers.

[1] L. Lam and C.Y. Suen, "A Theoretical Analysis of the Application of Majority Voting to Pattern Recognition," Proc. Int'l Conf. Pattern Recognition, IEEE CS Press, 1994, pp. 418-420.
[2] L. Xu, A. Krzyzak, and C.Y. Suen, “Methods of Combining Multiple Classifiers and Their Applications to Handwriting Recognition,” IEEE Trans. Systems, Man, and Cybernetics, vol. 22, no. 3, pp. 418-435, 1992.
[3] T.K. Ho, J.J. Hull, and S.N. Srihari, “Decision Combination in Multiple Classifiers Systems,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 16, no. 1, pp. 66-75, Jan. 1994.
[4] F. Kimura and M. Shridar, "Handwritten Numerical Recognition Based on Multiple Algorithms," Pattern Recognition, vol. 24, no. 10, pp. 969-983, 1991.
[5] C. Nadal, R. Legault, and C.Y. Suen, "Complementary Algorithms for the Recognition of Totally Unconstrained Handwritten Numerals," Proc. 10th Int'l Conf. Pattern Recognition, pp. 443-449,Atlantic City, N.J., 1990.
[6] E. Mandler and J. Schurmann, "Combining the Classification Results of Independent Classifiers Based on the Dempster-Shafer Theory of Evidence," Pattern Recognition and Artificial Intelligence, pp. 381-393,North Holland. Elsevier Science Publishers B.V., 1988.
[7] Y.S. Huang and C.Y. Suen, "A Method of Combining Multiple Classifiers—A Neural Network Approach," Proc. 12th Int'l Conf. Pattern Recognition and Computer Vision, pp. 473-475,Jerusalem, 1994.
[8] Y.S. Huang and C.Y. Suen, A Method of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 17, pp. 90-93, 1998.
[9] M. Sabourin, A. Mitiche, D. Thomas, and G. Nagy, "Classifier Combination for Handprinted Digit Recognition," Proc. Second Int'l Conf. Document Analysis and Recognition, pp. 163-166,Tsukuba Saenie City, Japan, Oct.20-22 1993.
[10] R.O. Duda and P.E. Hart, Pattern Classification and Scene Analysis. John Wiley and Sons, 1973.
[11] K. Knight, "Connectionist Ideas and Algorithms," Comm. ACM, vol. 33, no. 11, pp. 59-74, 1990.
[12] J.R. Quinlan, C4.5 Programs for Machine Learning. Morgan Kaufmann, 1992.
[13] P.A. Devijver and J. Kittler, Pattern Recognition: A Statistical Approach. Prentice Hall, 1982.
[14] W.P. Kegelmeyer Jr. and M.C. Allmen, "Dense Feature Maps for Detection of Calcifications," Digital Mammography: Proc. Second Int'l Workshop Digital Mammography, vol. 1,069, pp. 3-12, Int'l Congress Series, York, England. Elsevier Science B.V., July10-12 1994.
[15] C.E. Metz, "ROC Methodology in Radiologic Imaging," Investigative Radiology, vol. 21, pp. 720-733, 1986.
[16] J.A. Swets, "ROC Analysis Applied to the Evaluation of Medical Imaging Techniques," Investigative Radiology, vol. 14, pp. 109-121, 1979.
[17] J.A. Hanley and B.J. McNeil, "The Meaning and Use of the Area under a Receiver Operating Characteristic (ROC) Curve," Radiology, vol. 143, pp. 29-36, 1982.
[18] L. Breiman, J.H. Friedman, R.A. Olson, and C.J. Stone, Classification and Regression Trees.Belmont, Calif.: Wadsworth Int'l Group, 1984.

Index Terms:
Combination of classifiers, dynamic classifier selection, local classifier accuracy, classifier fusion, ROC analysis.
Kevin Woods, W. Philip Kegelmeyer Jr., Kevin Bowyer, "Combination of Multiple Classifiers Using Local Accuracy Estimates," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 4, pp. 405-410, April 1997, doi:10.1109/34.588027
Usage of this product signifies your acceptance of the Terms of Use.