One class classification with LibSVM in weka; unreasonable prediction??
I am doing a one class classification with the LibSVM classifier in weka.
I'm using "double pred = myClassi.classifyInstance(myInstance);" to get the predicted value. I tested with 6 instances from the same
training set and I get 4 instances as NaN and 2 as 0.0. I guess here NaN = outlier and 0.0 = relevant right?. But how do I get 4 outliers?
it's the same instances I used for training, so I should get all relevant. What could be the reason here? I didn't do any parameter tuning, only -S 2 to say unary classification. What's happening wrong here?
I really appreciate the help.
LibSVM in Weka is just a wrapper for the libsvm third-party library. I've never really used libsvm much because Weka's SMO always performs just as well (at least on the data sets I've tried). So I'm not much of an expert on libsvm - best to see if you can contact the libsvm guys and ask your question.