Table 6.
Ensemble learning classification evaluation performance for ensemble A and ensemble B with binary and multi-class identification scenarios. These results were recorded as an average for all respiratory classes.
| Models | Accuracy (ACC) | Sensitivity (SEN) | Specificity (SPE) | Precision (PRE) | F1-score | AUC |
|---|---|---|---|---|---|---|
| Binary Classification Results | ||||||
| Ensemble A | 0.9723 | 0.9722 | 0.9722 | 0.9724 | 0.9722 | 0.9722 |
| Ensemble B | 0.9644 | 0.9644 | 0.9644 | 0.9651 | 0.9644 | 0.9644 |
| Multi-class Classification Results | ||||||
| Ensemble A | 0.9720 | 0.9580 | 0.9790 | 0.9583 | 0.9580 | 0.9686 |
| Ensemble B | 0.9643 | 0.9464 | 0.9732 | 0.9493 | 0.9468 | 0.9598 |