Bagging and boosting algorithms for support vector machine classifiers

The performance of support vector machines (SVMs) greatly depends on the used values for hyperparameters. The tuning of hyper-parameters is a time-consuming task especially when the amount of data is large. In this paper, in order to overcome this difficulty, ensemble learning methods based on bagging and boosting are proposed. The proposed bagging methods reduce the computation time while keeping a reasonable accuracy. The proposed boosting method improves the accuracy of a conventional SVM classifier. The effectiveness of the proposed methods are demonstrated by numerical simulations.