Accession Number : AD0731221

Title :   K-Nearest Neighbor Rules.

Descriptive Note : Final phase rept.,

Corporate Author : PURDUE UNIV LAFAYETTE IND SCHOOL OF ELECTRICAL ENGINEERING

Personal Author(s) : Patrick,Edward A. ; Fischer,Frederic Philip , II

Report Date : SEP 1971

Pagination or Media Count : 170

Abstract : The k - Nearest Neighbor Rule is one specific example of a statistical decision rule. The report deals with a specific modification of this rule. The topics included are the distribution of risk of any sample - based decision rule in terms of the training set sample size, rule structure and underlying probability densities, evaluation of a k - nearest neighbor rule, the metric used in non parametric decision rules. (Author)

Descriptors :   (*STATISTICAL ANALYSIS, DECISION THEORY), (*PATTERN RECOGNITION, MATHEMATICAL PREDICTION), MULTIVARIATE ANALYSIS, DISTRIBUTION FUNCTIONS, SAMPLING, ARTIFICIAL INTELLIGENCE, VECTOR ANALYSIS, INTEGRAL EQUATIONS, RANDOM VARIABLES, CONVERGENCE, MATHEMATICAL MODELS, THESES

Subject Categories : Statistics and Probability
      Cybernetics
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE