Illustrative Examples of Clustering Using the Mixture Method and Two Comparable Methods from SAS.

Abstract

The technique of clustering uses the measurements on a set of element to identify clusters or groups of elements such that there is relative homogeneity within the groups and heterogeneity between the groups. In the associated technical report the mixture model approach is explained in detail and discussed in relation to other clustering techniques. Under this approach to clustering, the elements are assumed to be a sample from a mixture of several populations in various proportions. The practical application to two real data sets is considered here with the density function in each underlying population assumed to be normal. To provide a base for comparison, two SAS clustering methods with similar assumptions are also considered. The data are analysed using: KMM-Normal mixture model method, SAS (CLUSTER)- Ward's method, and SAS (CLUSTER)-Ward's method, and SAS (CLUSTER)-EML method. The results are discussed.

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 1987
Accession Number
ADA180503

Entities

People

  • K. E. Basford
  • N. J. Miles-mcdermott
  • W. T. Federer

Organizations

  • Cornell University

Tags

DTIC Thesaurus Topics

  • Clustering
  • Data Sets
  • Heterogeneity
  • Homogeneity

Fields of Study

  • Mathematics

Readers

  • Neural Network Machine Learning.
  • Regression Analysis.