R help - Jan 2008 - unsupervised random forest classification

Friends,
I would like to use Random Forest in unsupervised mode to classify data that are
from a mixture of categorical and continuous variables. The examples I find
online use only continuous data, followed by an mds plot. Is it correct to use
RF to obtain proximities for my mixed dataset and then perform some other form
of clustering (i.e. pam, clara, etc.) using the proximities matrix? Or is there
a way to perform unsupervised classification that will kick out cluster
membership on this dataset in RF itself?

Thanks,
Helen Poulos
Yale School of Forestry

Random Forests do not do clustering.  You need to take the proximity
matrix and feed it to algorithms of your choice for that.

Best,
Andy 

From: helen.mills at yale.edu
> 
> Friends,
> I would like to use Random Forest in unsupervised mode to 
> classify data that are
> from a mixture of categorical and continuous variables. The 
> examples I find
> online use only continuous data, followed by an mds plot. Is 
> it correct to use
> RF to obtain proximities for my mixed dataset and then 
> perform some other form
> of clustering (i.e. pam, clara, etc.) using the proximities 
> matrix? Or is there
> a way to perform unsupervised classification that will kick 
> out cluster
> membership on this dataset in RF itself?
> 
> Thanks,
> Helen Poulos
> Yale School of Forestry
> 
> ______________________________________________
> 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.
> 
> 
> 

------------------------------------------------------------------------------
Notice:  This e-mail message, together with any attachme...{{dropped:15}}