(also known as SVM) a supervised machine learning algorithm used for classification and regression tasks. It is a margin-maximising linear classifier. It tries to find the best boundary that separates different classes in the data.
Principle
In the study design, hyperplanes (the decision boundaries separating classes) are only one or two-dimensional. This means that the hyperplanes can only be 1D lines or 2D planes.
The margin is defined as a distance from the hyperplane to the nearest data points. The nearest data points to the line are called the support vectors.
The main goal of the algorithm is to maximise the margin between the two classes. The larger the margin, the better the model performs on new and unseen data. 1
Does not get stuck in local minima.
Creating a second feature from one-dimensional data
To distinguish data that is not linearly separable, we need to use a kernel trick (introducing a new feature)