Accession Number : AD0714810

Title :   Metric Considerations in Cluster Analysis.

Descriptive Note : Technical rept.,

Corporate Author : STANFORD UNIV CALIF DEPT OF STATISTICS

Personal Author(s) : Chernoff,Herman

Report Date : 20 SEP 1970

Pagination or Media Count : 20

Abstract : A variation of the 'k-means' method of cluster analysis is described which is designed to take into account and profit from the possibility that the separate clusters resemble samples from multivariate normal distributions with substantially different covariance structures. These covariance structures determine metrics which can be updated recursively. (Author)

Descriptors :   (*STATISTICAL ANALYSIS, CLASSIFICATION), PATTERN RECOGNITION, CORRELATION TECHNIQUES, MULTIVARIATE ANALYSIS, MEASURE THEORY, SAMPLING, THEOREMS

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE