Getting started with support vector machine
The SVM is a supervised classification method based in a kernel geometrical construction as is shown in the following figure. SVM can be applied either for classification or regression. SVM will look for the best decision boundary that split the points into the class that they belong. To accomplish this SVM, we will look for the largest margin (space that is free of training samples parallel to the decision boundary). In the following figure, we can see the margin as the space between the dividing line and dotted lines. SVM will always look for a global solution due to the algorithm only care about the vectors close to the decision boundary. Those points in the edge of the margin are the support vectors. However, this is only for two-dimensional spaces, when we have high-dimensional spaces the decision boundaries turn into hyperplane (maximum decision margin) and the SVMs will look for the maximum-margin hyperplanes. In this chapter we will only...
