Table 4 Performance of ML methods
From: Improvement of APACHE II score system for disease severity based on XGBoost algorithm
Abbreviation | Accuracy [95% Cl] | Precision [95% Cl] | Recall [95% Cl] | F1 [95% Cl] |
|---|---|---|---|---|
SVM | 0.736 ± 0.003 [0.732, 0.740] | 0.819 ± 0.005 [0.813, 0.825] | 0.736 ± 0.003 [0.732, 0.740] | 0.765 ± 0.002 [0.763, 0.767] |
LR | 0.704 ± 0.004 [0.698, 0.710] | 0.822 ± 0.003 [0.817, 0.826] | 0.704 ± 0.004 [0.698, 0.710] | 0.742 ± 0.003 [0.737, 0.746] |
NB | 0.841 ± 0.005 [0.834, 0.847] | 0.802 ± 0.007 [0.792, 0.812] | 0.841 ± 0.005 [0.834, 0.847] | 0.809 ± 0.004 [0.803, 0.814] |
ANN | 0.854 ± 0.002 [0.851, 0.858] | 0.827 ± 0.007 [0.817, 0.837] | 0.854 ± 0.002 [0.851, 0.858] | 0.810 ± 0.004 [0.805, 0.815] |
RF | 0.841 ± 0.003 [0.837, 0.845] | 0.840 ± 0.002 [0.837, 0.843] | 0.840 ± 0.001 [0.839, 0.841] | 0.841 ± 0.002 [0.838, 0.843] |
XGBoost | 0.858 ± 0.002 [0.855, 0.862] | 0.834 ± 0.005 [0.827, 0.841] | 0.858 ± 0.002 [0.855, 0.862] | 0.824 ± 0.003 [0.820, 0.829] |
Apache II | 0.742 ± 0.000 | 0.796 ± 0.000 | 0.742 ± 0.000 | 0.764 ± 0.000 |