The support vector machine is a supervised learning algorithmSupervised learning that performs two-class classification, and its emergence is one of the greatest milestones in the history of pattern recognition and machine learning. Supervised learning The basis of the support vector machine is the concept of the margin introduced in Chap. 2 . By formulating margin maximization as a convex quadratic programming problem, it is now possible to obtain discriminant functions with high generalization ability. Furthermore, by introducing the kernel trick introduced in Chap. 9 , support vector machines can be used in high-dimensional spaces, thus extending their range of application. Today, support vector machines have established themselves as a general-purpose pattern classification method that can be used with confidence for many problems, and have become one of the indispensable tools in all areas of information processing. In this chapter, we explain the functions and characteristics of support vector machines, taking into account their historical background.