Table 13. Classification results of various methods in terms of classification accuracy, sensitivity, specificity, number of selected features and support vectors, and model parameters for Australian Credit Approval (ACA).
| Criteria | Methods | ||||
|---|---|---|---|---|---|
| GAFS | PSOFS | TVPSOFS | FOASVM | CIFOA-SVM | |
| Sensitivity | 0.9429 + 0.0003 | 0.8289 + 0.0020 | 0.8152 + 0.0050 | 0.7552 + 0.0177 | 0.8376 + 0.0015 |
| Accuracy | 0.8200 + 0.0015 | 0.7991 + 0.0101 | 0.8231 + 0.0017 | 0.8115 + 0.0032 | 0.8376 + 0.0007 |
| Specificity | 0.5382 + 0.0093 | 0.8513 + 0.0057 | 0.8315 + 0.0042 | 0.8545 + 0.0052 | 0.8190 + 0.0018 |
| N of features | 8.1 + 3.69 | 8.3 + 6.01 | 6.5 + 2.35 | 14.0 + 0.0 | 6.3 + 1.29 |
| N of SVs | 540.8 | 402.2 | 399.5 | 384.8 | 403.2 |
| Parameter C | 105.1754 | 480.7929 | 439.4262 | 418.5625 | 401.1397 |
| Parameter γ | 0.5761 | 0.4315 | 0.4163 | 0.0399 | 0.5574 |