Accession Number : AD0680468

Title :   ADAPTIVE MODELLING OF LIKELIHOOD CLASSIFICATION.

Descriptive Note : Supplement to final rept.,

Corporate Author : PHILCO-FORD CORP BLUE BELL PA COMMUNICATIONS AND ELECTRONICS DIV

Personal Author(s) : Abend,Kenneth ; Harley,Thomas J. , Jr. ; Kanel,Laveen N.

Report Date : DEC 1968

Pagination or Media Count : 16

Abstract : An attempt was made to place the relationship between recognition and classification on the one hand and between theory and application of statistical classification on the other hand, in proper perspective. Compound decision theory is the latest step in the evolution of the most general model in which to imbed statistical classification problems arising in recognition system design. For the nonformalizable aspects of design, interactive approaches, namely those in which the human is part of the loop in the design process, with different classification and heuristic algorithms at his call, seem to be most promising. (Author)

Descriptors :   (*PATTERN RECOGNITION, DECISION THEORY), CLASSIFICATION, STATISTICAL ANALYSIS, AUTOMATIC, ARTIFICIAL INTELLIGENCE, READING MACHINES, MODEL THEORY

Subject Categories : Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE