Application and comparison of neural network, C5.0, and classification and regression trees algorithms in the credit risk evaluation problem (case study: a standard German credit dataset).