Research Article
Hyperparameter Tuning of Machine Learning Algorithms Using Response Surface Methodology: A Case Study of ANN, SVM, and DBN
Table 3
Best hyperparameter setting and the test performance of SVM.
| | Fold | Kernel function | C | Conv | MAE (validation) |
| | 1 | ANOVA | 1 | 0.0109 | 1.0668 | | 2 | Radial | 3 | 0.0307 | 1.1237 | | 3 | ANOVA | 2 | 0.0703 | 1.052 | | 4 | Radial | 75 | 0.001 | 1.0492 | | 5 | ANOVA | 1 | 0.1 | 1.061 | | 6 | Dot | −1 | 0.001 | 0.9971 | | 7 | ANOVA | 1 | 0.0406 | 1.1648 | | 8 | ANOVA | −1 | 0.0901 | 1.105 | | 9 | ANOVA | −1 | 0.0208 | 1.0693 | | 10 | ANOVA | 1 | 0.0802 | 1.1319 |
|
|
