Hi Erdal, Hi every body,
As far as I think we should know that what means actually a handy SVM. For clustering by SVM there is no any common software. LibSVM behinde MATLAB, R, and so on implements mainly classification, regression and density estimation and not clustering. Others like SVMlight, mySVM, kernlab do so. Thus, I think for clustering by SVM, Christian should search for a non-common software.
My own question which is like the question of Christian is that how it is possible to implement my own loss function by Spider for implementing SVM regression. Of course this question is answered at the website in FAQ/ optimization tool/ How do i implement my own loss function. But I want to know how optimization problem in SVM is affected by my own loss.
Thank you so much for any kind of response.
Amir
Hi Christian,
You may want to look at Spider software available at
http://www.kyb.tuebingen.mpg.de/bs/people/spider/
HTH
Erdal
>From: XN_UNM <[log in to unmask]>
>Reply-To: The Support Vector Machine discussion list
><[log in to unmask]>
>To: [log in to unmask]
>Subject: SVM software Question, thanks
>Date: Sun, 3 Jun 2007 22:37:14 -0700
>
>Hello All,
>
>Can anyone recommend a handy svm software that can do clustering jobs?
>Thanks a whole lot!
>
>Regards
>
>Christian
_________________________________________________________________
Spam filtresi ile virüslere karsi
en güvenilir koruma, MSN PC Koruma'dan
geçer. http://www.msn.com.tr/security/
Be a better Heartthrob. Get better relationship answers from someone who knows.
Yahoo! Answers - Check it out.