I'm looking for a dataset (preferably with a story, at any rate a real dataset) where a SVM with a linear kernel performs well...in other words i'm looking for a dataset where the class boundary is likely to be linear. Ideally, there should be between 10-20 continuous variables and not too many discrete ones....

It's not really important to me how many classes there are (as long as it's a classification problem). Also, it can't be the iris dataset.

Any proposals?

P.S.: @modo: i don't think this is a question about obtaining a particular dataset: it's more that i have not read svm/machine learning papers since i was a master student so i'm not familiar with some comon good examples...i'm just sure they exist.


How about this one http://archive.ics.uci.edu/ml/datasets/URL+Reputation ? It seems to be exactly what you are looking for.

  • 1
    $\begingroup$ Attribute Characteristics: Integer, Real $\endgroup$ – Diego Oct 9 '16 at 15:15
  • $\begingroup$ sorry I had misread this, $\endgroup$ – user603 Oct 9 '16 at 16:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.