Accession Number : AD0693825

Title :   PATTERN RECOGNITION IN THE ABSENCE OF A LEARNING PROCESS,

Corporate Author : FOREIGN TECHNOLOGY DIV WRIGHT-PATTERSON AFB OHIO

Personal Author(s) : Torgovitskii,I. Sh.

Report Date : 02 JAN 1969

Pagination or Media Count : 14

Abstract : Methods of pattern recognition in the absence of learning are examined. Information on the input situations enters the recognizer in the form of an n-dimensional vector so that any situation corresponds to a point in an n-dimension space. A situation is related to one of the classes according to the sign of the decision function. Given the equation of a hyperplane it is shown that, when certain factors are unknown, it is possible to determine the unknown coefficients that ensure the best division of the criterion space into two classes. Other methods are discussed. (Author)

Descriptors :   (*PATTERN RECOGNITION, DECISION THEORY), CLASSIFICATION, USSR

Subject Categories : Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE