Table 6 Performance evaluation of the selected ML algorithms for COVID-19 death prediction
From: Comparing machine learning algorithms for predicting COVID-19 mortality
Algorithms | Sensitivity (%) | Specificity (%) | Accuracy (%) | Precision (%) | ROC (%) |
|---|---|---|---|---|---|
Random forest | 90.70 | 95.10 | 95.03 | 94.23 | 99.02 |
XGBoost | 90.89 | 95.01 | 94.25 | 92.43 | 98.18 |
KNN | 97.38 | 82.15 | 89.56 | 80.11 | 96.78 |
MLP | 90.81 | 91.07 | 91.25 | 87.19 | 96.49 |
Logistic regression | 91.45 | 84.47 | 91.23 | 83.94 | 94.22 |
J48 decision tree | 87.77 | 94.47 | 92.17 | 89.97 | 92.19 |
Naïve Bayes | 90.44 | 84.31 | 87.47 | 81.32 | 92.05 |