# Some Studies on Pattern Recognition with Nonsupervised Learning Procedures

## Abstract

Generally speaking, the pattern recognition techniques must perform two basic functions, that is, the process of characterizing a class of the common pattern of inputs that belong to the same class and the process of classifying any input as a member of one of several classes. Now, the process of classification is based on a procedure of decision making in which a set of sample patterns may be attributed to the corresponding class. Let a set of sample patterns belonging to the same class be characterized by the set of parameters accompanied by the probability distribution which governs how each pattern involved in the same class is generated. This is the method of Parametric Statistics. In the parametric methods the training set whose statistical structures are in some cases known correctly and in other cases unknown is used for the purpose of obtaining estimates of the parameter values on the basis of the Bayes’ estimation theory, and the discriminant function which may implement the process of classification is then determined by these estimates. This is the so-called Bayes’ Machine.

## Keywords

Pattern Recognition Random Environment Posteriori Probability Sample Pattern Pattern Recognition Technique## Preview

Unable to display preview. Download preview PDF.

## References

- 1.H. J. Scudder, “Probability of Error of Some Adaptive Pattern-Recognition Machines,” IEEE Trans. on Information Theory, Vol. IT-11, No. 4, July 1965, p. 336.Google Scholar
- 2.E. M. Glaser, “Signal Detection by Adaptive Filters,” Trans. IRE, Vol. IT-7, No. 2, April 1961, p. 87.Google Scholar
- 3.N. Abramson, et.al., “Learning to Recognizer Patterns in a Random Environment,” Trans. IRE, Vol. IT-8, Sept. 1962, pp. 58–63.Google Scholar
- 4.F. Rosenblatt, “The Perceptron, A Perceiving and Recognizing Automaton,” Cornell Aeronautical Lab. Rept., No. 85-460-1, Jan. 1957.Google Scholar
- 5.B. Widrow, “Generalization and Information Storage in Networks of ADALINE “Neurons”, in Self-Organizing Systems edited by Yovits, Jacobi, and Goldstein (eds.), pp. 435–461, Spartan Books, Washington, D.C., 1962.Google Scholar
- 6.E. G. Henrichon and K. S. Fu, “A Nonparametric Partitioning Procedure for Pattern Classification,” IEEE Trans. on Computers, Vol. C-18, No. 7, July 1969, pp. 615–624.CrossRefGoogle Scholar
- 7.C. V. Jackowatz, R. L. Shuey and G. M. White, “Adaptive Waveform Recognition,” 4th Intl. Symp. on Inf. Theory, London, in Information Theory edited by C. Cherry, pp. 317–326, Butter-worths Book, 1961.Google Scholar
- 8.K. Tanaka, et.al., “An Identification Method of System Characteristics Using a New Type of Adaptive Correlating Filter,” Proc. IFAC, Tokyo Symp., August 1965, p. 245.Google Scholar
- 9.D. B. Cooper and P. W. Cooper, “Nonsupervised Adaptive Signal Detection and Pattern Recognition,” Information and Control, Vol. 7, No. 3, Sept. 1964, p. 416.CrossRefGoogle Scholar
- 10.K. S. Fu and C. H. Chen, “Sequential Decision, Pattern Recognition and Machine Learning,” Rept. TR-EE 65-6, Purdue Univ., April 1965.Google Scholar
- 11.C. G. Hilborn and D. G. Lainiotis, “Optimal Unsupervised Learning Multicategory Dependent Hypotheses Pattern Recognition,” IEEE Trans. on Information Theory, Vol. IT-14, May 1968, pp. 468–470.CrossRefGoogle Scholar
- 12.Y. T. Chien and K. S. Fu, “Stochastic Learning of Time-Varying Parameters in Random Environment,” IEEE Trans. on Systems Science and Cybernetics, Vol. SSC-5, No. 3, July 1969, pp. 237–246.CrossRefGoogle Scholar
- 13.K. Tanaka and S. Tamura, “Some Considerations on a Type of Pattern Recognition Using Nonsupervised Learning Procedure,” IFAC Intl. Symp. on Tech. and Biol. Problems of Control, Yerevan, Armenia, Sept. 1968. Also Trans., IECE of Japan, Vol. 52, No. 2, pp. 165–173, Feb. 1969.Google Scholar
- 14.S. Tamura and K. Tanaka, “On the Recognition of Time-Varying Patterns Using Learning Procedures”, IEEE Trans. on Information Theory, (to appear July 1971).Google Scholar
- 15.R. C. K. Lee, Optimal Estimation, Identification and Control, MIT Press, New York, 1969, pp. 43–49.Google Scholar