Hi, I am implementing KDDCUP dataset with SVM classification in MatLab by LIBSVM(https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/multilabel/).

Since KDDCUP is Multi-label binary classification does not support this.

As a solution, I used the MatLab tool from this URL:

https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/multilabel/

I ran this code :

[y, x, map] = read_sparse_ml('rcv1train.svm');

and now I've got the [y, x, map]. but I dont know where to use x, y, map in the svmclassification.

This code was an example for the binary classification:

matlab> [heart_scale_label, heart_scale_inst] = libsvmread('../heart_scale');

matlab> model = svmtrain(heart_scale_label, heart_scale_inst, '-c 1 -g 0.07');

but now I don't know what to do with [y, x, map] in this example.