Table 3.
Classification accuracy achieved using the optimal feature selection step.
| Classifier | Optimal Deep Model Features | Evaluation Measures | |||
|---|---|---|---|---|---|
| AlexNet Optimal | VGG16 Optimal | Accuracy (%) | Error Rate (%) | Time (Seconds) | |
| MC-SVM | ✓ | 96.2 | 3.8 | 14.277 | |
| ✓ | 94.2 | 5.8 | 15.004 | ||
| DT | ✓ | 90.1 | 9.9 | 15.167 | |
| ✓ | 91.2 | 8.8 | 17.286 | ||
| LDA | ✓ | 92.4 | 7.6 | 23.004 | |
| ✓ | 91.6 | 8.4 | 24.120 | ||
| KNB | ✓ | 92.7 | 7.3 | 45.115 | |
| ✓ | 90.3 | 9.7 | 47.016 | ||
| QSVM | ✓ | 93.9 | 6.1 | 17.336 | |
| ✓ | 94.8 | 5.2 | 19.224 | ||
| F-KNN | ✓ | 92.6 | 7.4 | 15.296 | |
| ✓ | 93.5 | 6.5 | 16.110 | ||
| Cosine KNN | ✓ | 93.4 | 6.6 | 15.804 | |
| ✓ | 94.9 | 5.1 | 16.299 | ||
| EBT | ✓ | 92.8 | 7.2 | 23.134 | |
| ✓ | 94.1 | 5.9 | 23.896 | ||