What is Support Vector Machines?

Support Vector Machines is a supervised learning algorithm to solve classification problem by finding the maximum-margin hyperplane.

Let’s say we have the wights and the bias for a label of panda. When we apply the wights and the bias on a panda image, it will have a very high score. On the other hand, when we apply other image (let’s say a dog image), the score is low. In order to distinguish a panda and others, we hope the score of a panda is higher by at least a margin. Let’s give an example.

We apply the wights and the bias on an unknown image. When the score is higher than the score of panda, it is classify as panda. However, we don’t wanna miss out some scores which can also classify as panda. Therefore we add a margin and have a wider range of score to classify the image as panda.