Not sure about your question, but if you want a classification model for labeling, firstly you need some data with labels for training, otherwise it is a clustering problem. If you want to use LibSVM, this file (https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf) will give you a very practical and helpful guide. Actually, this is where I started to learn and use SVM. Hope can help you.

I agree with answer (above), that "unlabeled data" is rather clustering. But perhaps for "unlabeled data" you mean, that all data have same label ? In this case you should use LibSVM with s = 2 (one-class SVM).

I attached a PDF file with some help.

Part of a MatLab example included in the LibSVM documentation is:

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

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

matlab> [predict_label, accuracy, dec_values] = svmpredict(heart_scale_label, heart_scale_inst, model); % test the training data

For probability estimates, you need '-b 1' for training and testing.

Hi

I used libsvm (training); now I want to use new data that has no dependent variable (Target). How can I do this?