An intuitive introduction to Support Vector Machine

Support Vector Machines (SVM), among classifiers, are probably the most intuitive and elegant, especially for binary classification tasks. To let you understand the intuition behind, I will explain them into a two-class environment: you will see that everything said will be valid also for multi classes tasks.

Let’s first consider the situation where our data are linearly separable.

Linearly Separable data

Generally speaking, the idea of SVM is finding a frontier which separates observations into classes. For this purpose, let’s consider the following example. Imagine you are a bank and you want to classify your clients among those who are creditworthy and those who are not, based on two features: monthly income and years spent working. To do so, you decide to collect some historical data of your past clients which have been already labeled as creditworthy or not creditworthy:

Our SVM algorithm is then asked to find a boundary which is able to segregate those two classes of clients, like so: