Table 3 Performance metrics of six algorithms after using Smote to handle data imbalance.

From: Machine learning for grading prediction and survival analysis in high grade glioma

Feature class

AUC

ACC

Recall rate

F1 score

Specificity

P

Logistic

0.81(0.77–0.84)

0.73

0.72

0.72

0.74

< 0.001

XGBoost

0.84(0.78–0.88)

0.78

0.79

0.77

0.77

< 0.001

Decision tree

0.77(0.66,0.84)

0.74

0.74

0.74

0.74

< 0.001

Random forest

0.86(0.82,0.92)

0.76

0.73

0.75

0.80

< 0.001

Adaboost

0.84(0.80,0.87)

0.76

0.74

0.75

0.77

< 0.001

GBDT

0.85(0.80,0.90)

0.76

0.76

0.76

0.76

< 0.001

Stacking

0.95(0.92–0.99)

0.85

0.84

0.85

0.87

 
  1. AUC Area under the ROC curve, ACC Accuracy, P values are used for stacking delong tests with other models.
Back to article page