Re: [R] svm and scaling input

From: Uwe Ligges <ligges_at_statistik.uni-dortmund.de>
Date: Wed 29 Jun 2005 - 05:30:43 EST

Arne.Muller@sanofi-aventis.com wrote:
> Dear All,
>
> I've a question about scaling the input variables for an analysis with svm (package e1071). Most of my variables are factors with 4 to 6 levels but there are also some numeric variables.
>
> I'm not familiar with the math behind svms, so my assumtions maybe completely wrong ... or obvious. Will the svm automatically expand the factors into a binary matrix? If I add numeric variables outside the range of 0 to 1 do I have to scale them to have 0 to 1 range?

Well, this depends on the kernel in use. For radial basis functions (as an example), you do not have to rescale, but a transformation of variables might make sense in order to get better results, though.

Uwe Ligges

>
> thanks a lot for help,
>
> +kind regards,
>
> Arne
>
> ______________________________________________
> R-help@stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Wed Jun 29 05:34:03 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:33:06 EST