Fig. 5.

Support vector machine. (A) In a two-dimensional SVM classification system, the maximum margin classifier is a straight line (red line). Support vectors are the nearest data points from the maximum margin classifier. The distance between support vectors and the maximum margin classifier is denoted as the soft margin. (B) In a two-group, one-dimensional data space, the decision boundary is a point, as shown by the red line. (C) In a two-group one-dimensional data space where the decision boundary cannot be drawn by a point, data are transformed by a kernel function to increase the dimension. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)