R help - Dec 2013 - Any way to pre-set the number of observations for each cluster with kmeans?

Hi Rusers,

This is a simple question, but I cannot find an answer to it yet.
I am currently running kmeans with a constraint that each cluster has at
least an N observations. I look at Kmeans and thought "nstart" is the
one,
but it didn't work.

Could you please let me know if there are other packages that will do this?

Thank you so much.

Best,

Taak

	[[alternative HTML version deleted]]

Hello Taak,

I don't believe that this can be done. What you are asking for is
essentially a constrained k-means algorithm with the constraint that
regardless of the consequences (higher within sum or squares), you
restrict each of the K groups to have at least N observations: within
this constraint, you would like to minimize the WSS. You have to
redevelop the k-means algorithm and rewrite the appropriate code as
needed.

It appears that you may have a specific application for which you need
this particular set-up. Do you really need to use k-means for this
clustering? 

Many thanks,
Ranjan

On Sun, 29 Dec 2013 19:57:56 +0900 Takatsugu Kobayashi
<taquito2007 at gmail.com> wrote:
> Hi Rusers,
> 
> This is a simple question, but I cannot find an answer to it yet.
> I am currently running kmeans with a constraint that each cluster has at
> least an N observations. I look at Kmeans and thought "nstart" is
the one,
> but it didn't work.
> 
> Could you please let me know if there are other packages that will do this?
> 
> Thank you so much.
> 
> Best,
> 
> Taak
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 

-- 
Important Notice: This mailbox is ignored: e-mails are set to be
deleted on receipt. Please respond to the mailing list if appropriate.
For those needing to send personal or professional e-mail, please use
appropriate addresses.

____________________________________________________________
TRY FREE IM TOOLPACK at http://www.imtoolpack.com/default.aspx?rc=if5
Capture screenshots, upload images, edit and send them to your friends
through IMs, post on Twitter?, Facebook?, MySpace?, LinkedIn? ? FAST!