[R] decision trees using the Hellinger distance rather than
eiverson at NMDP.ORG
Fri Oct 2 17:43:46 CEST 2009
Do you happen to have a large .Rdata file that is being loaded, or something in your .Rprofile? Try searching for a file with that name. Or start R with a --vanilla and see if that helps...
> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
> On Behalf Of Rajarshi Guha
> Sent: Friday, October 02, 2009 10:33 AM
> To: R
> Subject: [R] decision trees using the Hellinger distance rather than
> Hi, while working with decision trees and unbalanced data, I came across
> use of the Hellinger distance as an alternative to information gain [1,2],
> when dealing with skewed data. Does anybody know of R implementations of
> this approach to decision trees?
>  http://www.cse.nd.edu/Reports/2008/TR-2008-06.pdf
>  http://csmr.ca.sandia.gov/~wpk/slides/wdmda-sem.pdf
> Rajarshi Guha
> NIH Chemical Genomics Center
> [[alternative HTML version deleted]]
> R-help at r-project.org mailing list
> PLEASE do read the posting guide http://www.R-project.org/posting-
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help