[R] Variable importance in Random Forests

Liaw, Andy andy_liaw at merck.com
Thu Mar 6 17:34:07 CET 2008


Depending on the way you ran randomForest, the variable of interest may
have been used in many splits in many trees.  The information can be
extracted using the getTree() function.  However, I suspect it might be
more useful to look at partialPlot().

Best,
Andy 

> -----Original Message-----
> From: r-help-bounces at r-project.org 
> [mailto:r-help-bounces at r-project.org] On Behalf Of Nagu
> Sent: Wednesday, March 05, 2008 6:31 PM
> To: r-help at r-project.org; r-help at stat.math.ethz.ch
> Subject: [R] Variable importance in Random Forests
> 
> Hi,
> 
> How do I get what splits were used for the top important variables.
> For example, if my top most important variable is a continuous
> variable, e.g. previous 12 month sales, and its range is between $0 to
> $1M, how do I find what what were the split values for this variable
> that were used in the random forests algorithm? How do I achieve the
> same thing for a factor variable?
> 
> Thank you,
> Nagu
> 
> ______________________________________________
> 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}}



More information about the R-help mailing list