Support Vector Machine classifiers can best be understood geometrically. In
the simplest case, consider a set of points in a two-dimensional plane, some
belonging to class A, and some belonging to class B.We are given a training set of
points whose class (Aor B) is known, and we need to build a classifier of points,
using these training points. This situation is illustrated in Figure 20.5, where the
points in class A are denoted by X marks, while those in class B are denoted by O
marks.