Table 3.
Comparative evaluation of the performance of various ML models.
| Model | Sensitivity (%) | Specificity (%) | AUC | Accuracy (%) | Accuracy (weighted)[%] | MCC (%) |
|---|---|---|---|---|---|---|
| MLP Classifier | 91.3 | 87.5 | 89.3 | 89.1 | 88.9 | 78.3 |
| Ada Boost Classifier | 85.8 | 66.6 | 76.3 | 75 | 76.2 | 52.5 |
| Logistic Regression | 77.1 | 74.1 | 75.6 | 75.5 | 75.3 | 50.9 |
| SVC | 84.7 | 65.8 | 75.2 | 74.1 | 75.2 | 50.5 |
| Cat Boost Classifier | 79.3 | 68.3 | 73.8 | 73.1 | 73.5 | 47.3 |
| Extra Trees Classifier | 77.1 | 65 | 71.1 | 70.3 | 70.8 | 41.9 |
| XGB Classifier | 79.3 | 58.3 | 68.9 | 67.4 | 69 | 37.8 |
| Random Forest Classifier | 75 | 60 | 67.6 | 66.5 | 67.4 | 34.9 |
Abbreviations: AUC: area under the curve; MCC: Matthews's correlation coefficient; MLP: Multiple Perceptron (MLP); ML: machine learning.