Fig. 1.
Illustration of the concept used in support vector machines. The algorithm tries to find a boundary that maximizes the distance (d) between groups. The figure reduces the problem to two groups and two dimensions for the purpose of illustration only.