[R] interpretation of svm models with the e1071 package

[Steve Lianoglou]

Hi,

On Mon, Jul 12, 2010 at 4:55 AM, manuel.martin
<manuel.martin at orleans.inra.fr> wrote:
> On 07/10/2010 04:11 AM, Steve Lianoglou wrote:
>> On Fri, Jul 9, 2010 at 12:15 PM, manuel.martin
>> <manuel.martin at orleans.inra.fr>  wrote:
<snip>
>>
>>>
>>> Dear all,
>>>
>>> after having calibrated a svm model through the svm() command of the
>>> e1071
>>> package, is there a way to
>>> i) represent the modeled relationships between the y and X variables
>>> (response variable vs. predictors)?
>>>
>>
>> Can you explain a bit more ... how do you want them represented?
>>
>
> I was thinking to a simple ŷ = fi(Xi) plot, fi resulting from the fitted svm
> model. Xi is the predictor, among the whole set of predictors, X, one wish
> to see the relationship with the response.
> For boosted regression trees, which I am more familiar with, this is fi
> function is estimated by averaging the effects of all predictors but Xi, and
> plotting how ŷ varies as Xi does.

I still think you might be able to get some mileage out of calculating
your W vector and looking at the values in each of its
coordinates/bins.

I think one problem trying to figure out something for the plot you
are after is that I feel like depending on the choice of kernel used
in for your SVM, rigging up such an fi(Xi) plot might not be as
straight forward as you might think, since kernels can manipulate your
feature space in fun ways.

There's some literature out there about how to extract
meaning/features from an SVM model. Perhaps you can search through
some of that to help get some ideas.

-- 
Steve Lianoglou
Graduate Student: Computational Systems Biology
 | Memorial Sloan-Kettering Cancer Center
 | Weill Medical College of Cornell University
Contact Info: http://cbio.mskcc.org/~lianos/contact