Based on my very simple understanding of SVMs, it seems like a probabilistic output would be a very useful thing to have. Soft margin seems to part of the way toward accounting for noisy data, but still classifies it into one class or another without saying anything about how likely a given point is to be classified correctly. I've found a couple proposed ways to get this with a simple Google search, but am curious what is the state-of-the-art and what is the most straight forward way of getting this probabilistic output. Thanks



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.