Modified matrix splitting method for the support vector machine and its application to the credit classification of companies in Korea

► We propose a solving approach for the ν-support vector machine (SVM) for classification problems. We use the modified matrix splitting method and incomplete Cholesky decomposition. The matrix splitting method combined with the projection gradient method. The incomplete Cholesky decomposition is used for the large scale Hessian. The proposed method applies for the credit prediction for small-sized Korean companies.