Improving Quality of the Multiclass SVM Classification Based on the Feature Engineering

Abstract

The SVM classifier is effective in solving binary classification problems. However, in practical problems of classification, there are often cases of the presence of more than two classes of objects in the original data set. This work is devoted to the study of approaches to improving the quality of the SVM classification based on the engineering of new features of objects in the original data set using tools of such well-known multiclass classification algorithms as the Decision Tree, Random Forest and AdaBoost.