Accession Number : AD0720837

Title :   Pattern Recognition with Continous Parameter, Observable Markov Chains.

Descriptive Note : Interim scientific rept. no. 10,

Corporate Author : MICHIGAN STATE UNIV EAST LANSING DIV OF ENGINEERING RESEARCH

Personal Author(s) : Dubes,Richard C. ; Panayirci,Erdal

Report Date : 25 NOV 1970

Pagination or Media Count : 34

Abstract : The paper develops Bayesian learning and decision-making algorithms for the following pattern recognition problem. Each of M pattern classes is described by a continuous-parameter, discrete-state Markov chain having a finite number of states. All states and times of transition between states can be observed perfectly. The transition rate matrices, which establish the properties of the chains, are not known a priori. A Bayesian learning algorithm using a fixed amount of memory digests the training patterns which consist of a member function from each chain. This leads to an iterative, computationally simple, decision-making algorithm for classifying any portion of a member function. The Bhattacharyya bound and the probability of error are derived for the 2-state, 2-chain problem when the transition rate matrices are known. The last section reports on a computer simulation of a 3-state, 2-chain problem with varying amounts of training data. An appendix summerizes the pertinent facts about Markov chains. (Author)

Descriptors :   (*PATTERN RECOGNITION, *DECISION THEORY), STOCHASTIC PROCESSES, DECISION MAKING, ELECTROENCEPHALOGRAPHY, MATRICES(MATHEMATICS), ERRORS, ALGORITHMS

Subject Categories : Operations Research
      Cybernetics
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE