ABSTRACT
The support vector machine (SVM) has been introduced to various pattern classification and function approximation problems. Pattern classification is used to classify some objects into one given category called classes. Hard margin SVM can work only when data is completely linearly separable without any errors. In case of errors, either the margin is smaller or the hard margin SVM fails. SVM using direct decision functions, an extension of multi-class problems, is not straightforward. Followings are different types of SVM that handle multi-class problems. The SVM has been applied to the problem of predictions and proven to be superior to competing methods, such as the neural network, the linear multiple discriminant approaches, and logistic regression. Decision tree–based SVM, which combines support vector machines and decision tree, is an effective way for solving multi-class problems. The robust SVM aims to solve the overfitting problem with outliers that make the two clas.
