Hematoma expansion prediction based on SMOTE and XGBoost algorithm

Table 5 The predictive results on different time groups

Methods	Accuracy	Precision	Recall	F1-score
(a) Time group T₁
XGBoost	0.89	0.91	0.90	0.89
SVM [1]	0.77	0.84	0.76	0.75
RF [2]	0.86	0.89	0.89	0.88
LR [14]	0.74	0.78	0.75	0.74
KNN [15]	0.82	0.88	0.88	0.88
(b) Time group T₂
Methods	Accuracy	Precision	Recall	F1-score
XGBoost	0.82	0.83	0.83	0.80
SVM [1]	0.70	0.70	0.83	0.69
RF [2]	0.80	0.81	0.81	0.78
LR [14]	0.69	0.69	0.67	0.68
KNN [15]	0.78	0.78	0.77	0.78
(c) Time group T₃
Methods	Accuracy	Precision	Recall	F1-score
XGBoost	0.87	0.87	0.88	0.87
SVM [1]	0.77	0.82	0.74	0.74
RF [2]	0.85	0.85	0.85	0.85
LR [14]	0.74	0.74	0.75	0.74
KNN [15]	0.86	0.86	0.87	0.87
(d) Time group T₄
Methods	Accuracy	Precision	Recall	F1-score
XGBoost	0.94	0.94	0.93	0.93
SVM [1]	0.79	0.85	0.80	0.78
RF [2]	0.92	0.92	0.92	0.92
LR [14]	0.84	0.84	0.84	0.84
KNN [15]	0.92	0.92	0.92	0.92

ISSN: 1472-6947