Table 12. Classification results of various methods in terms of classification accuracy, sensitivity, specificity, number of selected features and support vectors, and model parameters for German Credit Data (GCD).
| Criteria | Methods | ||||
|---|---|---|---|---|---|
| GAFS | PSOFS | TVPSOFS | FOASVM | CIFOA-SVM | |
| Sensitivity | 0.9213 + 0.0026 | 0.9246 + 0.0006 | 0.9359 + 0.0008 | 0.9282 + 0.0006 | 0.9503 + 0.0007 |
| Accuracy | 0.8150 + 0.0014 | 0.7980 + 0.0013 | 0.8060 + 0.0016 | 0.8160 + 0.0019 | 0.8170 + 0.0006 |
| Specificity | 0.5524 + 0.0009 | 0.5027 + 0.0105 | 0.5059 + 0.0018 | 0.5548 + 0.0015 | 0.5104 + 0.0008 |
| N of features | 17.1 + 6.49 | 15.3 + 28.1 | 16.4 + 23.4 | 24.0 + 0.0 | 16.1 + 3.09 |
| N of SVs | 540.965 | 557.9 | 527.6 | 518.1 | 504.3 |
| Parameter C | 161.4932 | 166.4563 | 146.5901 | 747.4506 | 166.3319 |
| Parameter γ | 0.3754 | 0.6177 | 0.5136 | 0.0287 | 0.6666 |