Fig. 4

Sample points on the hyper plane (also referred to as support vectors). Obviously, if f(x) = 0, then x is the point on the hyper plane. It may be required that for all points that satisfy f(x) < 0, the corresponding y is equal to y = −1, and f(x) > 0 corresponds to data points where y = 1. wx denotes the inner product of w and x in the classification function f(x) = wx + b. In the common sense, w is the normal vector (weight vector) and b is the intercept (bias)