Parameter tuning of large scale support vector machines using ensemble learning with applications to imbalanced data sets

Parameter tuning for kernels affects the generalization ability of support vector machine (SVM). Although the cross validation method is widely applied to this aim, it is usually time consuming. This paper applies ensemble learning using both Bagging and Boosting to parameter tuning in SVM. It will be shown that the proposed method is effective in particular for large scale data sets and for imbalanced data sets.