US and Worldwide: +1 (866) 660-7555
Results 1 to 3 of 3

Thread: One class classification with LibSVM in weka; unreasonable prediction??

  1. #1
    Join Date
    Apr 2012
    Posts
    20

    Default One class classification with LibSVM in weka; unreasonable prediction??

    Hi,
    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.
    Thanks.

  2. #2
    Join Date
    Aug 2006
    Posts
    1,424

    Default

    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.

    Cheers,
    Mark.

  3. #3
    Join Date
    Apr 2012
    Posts
    20

    Default

    ok I will. thanks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •