Accession Number : AD0667728

Title :   ON PATTERN CLASSIFICATION ALGORITHMS--INTRODUCTION AND SURVEY.

Descriptive Note : Technical rept.,

Corporate Author : HARVARD UNIV CAMBRIDGE MASS DIV OF ENGINEERING AND APPLIED PHYSICS

Personal Author(s) : Ho,Yu-Chi ; Agrawala,Ashok K.

Report Date : MAR 1968

Pagination or Media Count : 52

Abstract : This paper attempts to lay bare the underlying ideas used in various pattern classification algorithms reported in the literature. It is shown that these algorithms can be classified according to the type of input information required and that the techniques of estimation, decision, and optimization theory can be used to effectively derive known as well as new results. (Author)

Descriptors :   (*PATTERN RECOGNITION, ALGORITHMS), (*ARTIFICIAL INTELLIGENCE, *LEARNING MACHINES), CLASSIFICATION, DECISION THEORY, OPTIMIZATION, DIGITAL COMPUTERS, ITERATIONS, PROBABILITY, MINIMAX TECHNIQUE, SEQUENCES(MATHEMATICS), MATRICES(MATHEMATICS), ADAPTIVE SYSTEMS

Subject Categories : Computer Hardware
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE