[prev in list] [next in list] [prev in thread] [next in thread] 

List:       wekalist
Subject:    Re: [Wekalist] General Question on EM Clustering
From:       Eibe Frank <eibe () waikato ! ac ! nz>
Date:       2012-11-29 21:54:04
Message-ID: 1F9DE930-C652-41B8-BDB9-C5B01B6ECEAA () waikato ! ac ! nz
[Download RAW message or body]

Take a look at the following paper by Padhraic Smyth:

  http://www.aaai.org/Papers/KDD/1996/KDD96-021.pdf

Cheers,
Eibe

On 30 Nov 2012, at 10:35, Jeff Pattillo wrote:

> The book "Data Mining" associated with Weka claims the EM implementation in Weka
> can either have the number of clusters specified by the user, or can decide
> using cross-validation.  
> 
> I'm trying to picture how cross-validation could be used to decide the number of
> clusters, since there is no supervisor when clustering.  Can someone either
> explain this generally or refer me to a paper on this?  Does anyone know the
> method Weka employs to find the number of clusters via cross-validation and/or
> the paper it comes from?
> 
> Thanks in advance.
> 
> Jeff
> 
> 
> _______________________________________________
> Wekalist mailing list
> Send posts to: Wekalist@list.scms.waikato.ac.nz
> List info and subscription status: https://list.scms.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html




_______________________________________________
Wekalist mailing list
Send posts to: Wekalist@list.scms.waikato.ac.nz
List info and subscription status: https://list.scms.waikato.ac.nz/mailman/=
listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.=
html


[prev in list] [next in list] [prev in thread] [next in thread]