Support vector machine is a machine learning algorithm based on statistical learning theory presented by Vapnik (1998) [66], which uses a non-linear transformation to map the input data to a high dimensional feature space where linear classification is performed. It is equivalent to solving the quadratic optimization problem: