| | This Article | |
| |
| |
| | Share | |
| |
| |
| | Bibliographic References | |
| |
| |
| | Add to: | |
| |
 Digg
 Furl
 Spurl
 Blink
 Simpy
 Google
 Del.icio.us
 Y!MyWeb
| |
| | Search | |
| |
| |
| | |
Effects of Sample Size in Classifier Design
August 1989 (vol. 11 no. 8)
pp. 873-885
The effect of finite sample-size on parameter estimates and their subsequent use in a family of functions are discussed. General and parameter-specific expressions for the expected bias and variance of the functions are derived. These expressions are then applied to the Bhattacharyya distance and the analysis of the linear and quadratic classifiers, providing insight into the relationship between the number of features and the number of training samples. Because of the functional form of the expressions, an empirical approach is presented to enable asymptotic performance to be accurately estimated using a very small number of samples. Results were experimentally verified using artificial data in controlled cases and using real, high-dimensional data.
[1] 873T. S. El-Sheikh and A. G. Wacker, "Effect of dimensionality and estimation on the performance of Gaussian classifiers,"Pattern Recognition, vol. 12, pp. 115-126, 1980.[2] A. K. Jain and B. Chandrasekaran, "Dimensionality and sample size considerations in pattern recognition practice," inHandbook of Statistics, vol. 2, P. R. Krishnaiah and L. N. Kanal, Eds, Amsterdam, The Netherlands: North-Holland, 1982, pp. 835-855.[3] S. Raudys and V. Pikelis, "On dimensionality, sample size, classification error, and complexity of classification algorithm in pattern recognition,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI- 2, no. 3, pp. 242-252, May 1980.[4] C. P. Han, "Distribution of discriminant function in circular models,"Inst. Star. Math. Ann., vol. 22, no. 1, pp. 117-125, 1970.[5] G. J. McLachlan, "Some expected values for the error rates of the sample quadratic discriminant function,"Australian J. Stat., vol. 17, no. 3, pp. 161-165, 1975.[6] H. V. Pipberger, "Computer analysis of electrocardiogram," inClinical Electrocardiography and Computers, C. A. Caceres and L. S. Dreifus, Eds. New York: Academic, 1970, pp. 109-119.[7] A. K. Jain, "On an estimate of the Bhattacharyya distance,"IEEE Trans. Syst., Man, Cybern., pp. 763-766, Nov. 1976.[8] H. M. Kalayeh and D. A. Landgrebe, "Predicting the required number of training samples,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-5, no. 6, pp. 664-667, Nov. 1983.[9] D. M. Foley, "Considerations of sample and feature size,"IEEE Trans. Inform. Theory, vol. IT-18, pp. 618-626, 1972.[10] K. Fukunaga,Introduction to Statistical Pattern Recognition. New York: Academic, 1972.[11] L. Novak, "On the sensitivity of Bayes and Fisher classifiers in radar target detection," inProc. 18th Asilomar Conf. Circuits, Systems, and Computers, Nov. 5-7, 1984.[12] W. Beyer,CRC Standard Mathematical Tables, 26th ed. Boca Raton, FL: CRC Press, 1981, pp. 44-45.
Index Terms:
pattern recognition; classifier; design; sample-size; parameter estimates; bias; variance; Bhattacharyya distance; parameter estimation; pattern recognition
Citation:
K. Fukunaga, R.R. Hayes, "Effects of Sample Size in Classifier Design," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, no. 8, pp. 873-885, Aug. 1989, doi:10.1109/34.31448
