Are there any package that is using "v-fold cross-validation algorithm" to test the number the clusters?
Thanks, Ming On Sat, Mar 19, 2005 at 12:44:42PM +0100, Uwe Ligges wrote: > XP Sun wrote: > > > hi, all, > > > > how to decide the number of cluster before you use kmeans and hclust? > > thank you in advance! > > Depends on your criterion. Best idea is always to use the brain and > think about how many clusters are sensible for the particular > task/problem/data. > For hclust, you can also look at the dendrogram's height and distances > of dissimilarities in order to cut. > > Uwe Ligges > > > > best > > -xpsun > > > > ______________________________________________ > > [email protected] mailing list > > https://stat.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! > > http://www.R-project.org/posting-guide.html > > ______________________________________________ > [email protected] mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html ______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
