[R] Quick question on kmeans

[Gavin Simpson]

On Tue, 2008-01-15 at 14:34 +0100, Serguei Kaniovski wrote:
> Hello,
> 
> when kmeans draws random vectors for the initial centroids, does it then 
> select the clustering that has emerged most frequently?

No. It simply selects random starting centres and converges to a
solution, which may be a local not a global minimum. You'll have to do
the checking yourself to see if similar clusterings are obtained from
repeated calls to kmeans.

G

> 
> Serguei

-- 
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%
 Dr. Gavin Simpson             [t] +44 (0)20 7679 0522
 ECRC, UCL Geography,          [f] +44 (0)20 7679 0565
 Pearson Building,             [e] gavin.simpsonATNOSPAMucl.ac.uk
 Gower Street, London          [w] http://www.ucl.ac.uk/~ucfagls/
 UK. WC1E 6BT.                 [w] http://www.freshwaters.org.uk
%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%