Table 2 Performance metrics for six algorithms that do not use Smote to handle data imbalance.
From: Machine learning for grading prediction and survival analysis in high grade glioma
Feature class | AUC (95% CI) | ACC | Recall rate | F1 score | Specificity |
|---|---|---|---|---|---|
Logistic | 0.74(0.65–0.83) | 0.73 | 0.44 | 0.51 | 0.86 |
XGBoost | 0.75(0.67–0.83) | 0.74 | 0.49 | 0.54 | 0.86 |
Decision tree | 0.66(0.59–0.75) | 0.72 | 0.51 | 0.53 | 0.82 |
Random forest | 0.76(0.69–0.83) | 0.72 | 0.32 | 0.42 | 0.90 |
Adaboost | 0.71(0.64–0.79) | 0.72 | 0.44 | 0.50 | 0.85 |
GBDT | 0.75(0.68–0.81) | 0.73 | 0.49 | 0.53 | 0.84 |
