KM – K-Means Clustering of Cases

K-Means Clustering of Cases using BMDP Statistical Software Program KM

KM partitions cases into clusters with the result that each case belongs to a cluster whose center is closest to the case. KM standardizes the data and begins with all data in one cluster or with user-specified clusters and at each step reallocates cases to the closest (via euclidean distance) cluster. For each variable at each step, KM looks for the split with would reduce the within cluster variance the most and across all variables splits the cluster using this variable. Km is useful for large data sets, and it supplies infromation about the role of individual variables in the clustering, and it provides graphical displays.

• 4 Standardization methods
• Ability to specify initial cluster membership and save final cluster membership
• ANOVA, cluster profile, covariances and correlations
• Cross tabulation of user-selected variables versus final cluster membership
• View Program BMDP Program KM Manual Chapter