A method of supervised pattern recognition by an adaptive hypersphere decision threshold

Author

Darroll Steven McCormack

Abstract

"In this study the Bayes likelihood detector is combined with an adaptive decision threshold classifier to solve the multicategory pattern recognition problem. It is assumed that the pattern classes can be represented by an n-dimensional vector sample taken from a multivariate gaussian probability distribution. This study presents (1) the derivation of the A̲daptive H̲ypersphere D̲ecision T̲hreshold classifier (AHDT classifier) and shows (2) how the AHDT classifier minimizes the probability of error using the learning patterns. Finally the AHDT classifier is applied to the solution of a physical problem through computer simulation"--Abstract, page ii.

Advisor(s)

Kern, Frank J.

Committee Member(s)

Bertnolli, Edward C.
Antle, Charles E.

Department(s)

Electrical and Computer Engineering

Degree Name

M.S. in Electrical Engineering

Publisher

University of Missouri at Rolla

Publication Date

1968

Pagination

xi, 104 pages

Rights

© 1968 Darroll S. McCormack, All rights reserved.

Document Type

Thesis - Open Access

File Type

text

Language

English

Subject Headings

Adaptive control systemsPattern perceptionPattern recognition systems

Thesis Number

T 2136

Print OCLC #

5998338

Electronic OCLC #

794240951

Recommended Citation

McCormack, Darroll Steven, "A method of supervised pattern recognition by an adaptive hypersphere decision threshold" (1968). Masters Theses. 5190.
https://scholarsmine.mst.edu/masters_theses/5190