cluster seed for the k-means procedureWhat I am trying to do is to read in cluster seed for the k-means procedure. As you know K-means is quite sensitive to starting points and I have it found it useful to read in cluster means from a hierarchical solution. I have done this in SAS using PROC FASTCLUS and PROC CLUSTER in tandem and found it to be very useful.
I was trying to do the same thing in JMP. I was able to save the cluster means from hierarchical clustering into a data table, but could not read this as starting seed for K-means. Wondering if there is a way to do this in JMP?
Statistics and AnalyticsRe: cluster seed for the k-means procedure - Status changed to: AcknowledgedRe: cluster seed for the k-means procedureThere's no facility for this currently. We'll see if we can add something for this in a future release.
