Classification With Support Vector Machines, New Quadratic Programming Algorithm

Support vector machines (SVM) are excellent tools for classification and regression. They seek the optimal separating hyperplan and maximal margin. The modeling results often lead to solving a quadratic programming problem. In this paper, we present a simple method to determine the hyperplan H that separates two classes of examples so that the distance between these two classes is maximal. This method is based on the geometric interpretation of the norm of a linear mapping. The result model of our algorithm modeling is a maximization of a concave quadratic program. This quadratic program is resolved by projection method. Example illustrates the method.