Hello,

You can use the Libsvm: http://www.csie.ntu.edu.tw/~cjlin/libsvm/

It is free and a matlab interface is available.

HI,

you can also give a look to theKernel Machine webpage:

http://www.kernel-machines.org/

Regards,

Matteo

look at msvm 

http://www.loria.fr/~lauer/MSVMpack/MSVMpack.html

http://www.codeforge.com/s/0/MSVM

My experience with libsvm was good. So you probably can start from there.

Training is your data in column vector with each column as your feature, and groups represent your label(class of that row i,e one entity feature set) for each data.. Number of rows in Training and Groups should be same. 

Training with two features:    groups

feature1    feature 2            

5                  2                      

3                 5          

Label vector:

1

2..

I hope it helps.. I have two features in my feature matrix and for each row i have one label. Set your data accordingly.

Better is to use Libsvm for multiclass.. It is faster and efficient. Just take care about inversion in passing parameters. libSvm : For training use this comment. It will save your trained model and you can use this for testing any point of time.

tic

svmModel = svmtrain(groups, training_instance_matrix, ['-t 0 -c 1 ']);

toc

%# test using test instances

save('svmtrainfile.mat','svmModel');

After this you can test: 

load svmtrainfile;

[predicted_label, accuracy, decision_values] = svmpredict(testing_label_vector, tempmat, svmModel );

 Hugs

ive used a version of this software for 2 classes. works very well. multiclass code can be downloaded here. http://svmlight.joachims.org/

Refer to the link

https://stackoverflow.com/.../full-example-of-multiple-class-svm-with-cross-validation...