[R] Obtaining the surrogate splits from rpart

Clayton Springer csprin at brandybuck.ca.sandia.gov
Wed Oct 31 20:03:46 CET 2001


On Tue, Oct 30, 2001 at 07:51:59PM +0000, Prof Brian Ripley wrote:
> On Tue, 30 Oct 2001, Clayton Springer wrote:
> 
> > Dear r-help,
> >

> > > data(kyphosis)
> > > fit <- rpart(Kyphosis ~ Age + Number + Start, data=kyphosis)
> 
> summary.rpart prints them out for you, so just examine its R code
> to see how to extract them.

I did that and found that only some of the surrogate splits are available.
I would like to have the surrogate split for every feature at each node.
Will I be able to print this out from an rpart object?  
Are all the surrogate splits even calculated in rpart, it doesn't seem like they are.

Has anybody found all the surrogate splits and/or calculated Brieman's feature rankings? 

thanks again,


Clayton

-- 
Clayton Springer, Ph. D.
Sandia National Laboratories                      csprin at ca.sandia.gov
Biosystems Research Department                    (925) 294-2143
P.O. Box 969, MS 9951                fax:         (925) 294-3020
Livermore, CA, 94551-0969

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list