Empty clusters in k-means - possible solution

Dear Simon,

I respectfully disagree with you, when you say that it is NOT a bug. Empty
clusters can only happen if one provides a set of centers; if one enters
something like k=4, it cannot happen, so it must be a bug. Which one is
your case?

From the help:

"Except for the Lloyd-Forgy method, k clusters will always be returned if a
number is specified.  If an initial matrix of centres is supplied, it is
possible that no point will be closest to one or more centres, which is
currently an error for the Hartigan-Wong method"

This having been said, I suspect that there is a bug in the current
implementation of kmeans. I get the "empty cluster: try ..." answer even
when I enter a number of clusters, but only with R 2.15. I never had it
with R 2.14. Didn't try with R 3, yet.
I posted this problem on the list twice, at no avail (unless I forgot to
find the answer in the mailing list, of course).

Kind regards,
Luca

Hello all,

k-means algorithms can at times fail because one of the cluster become
emmpty. In this case, the kmeans R function returns:
"empty cluster: try a better set of initial centers"

This has been discussed several times on several R-lists, and is NOT a
bug, but can be annoying when using k-means in complex simulation where
this error brings everything to a stop. One can use try() or tryCatch()
to avoid this, but this is just a programming trick.

--