Tree-Based Risk Factor Identification and Stroke Level Prediction in Stroke Cohort Study

Table 3

Performance evaluation using fivefold cross-validation for different models: mean (standard deviation).

Model	AUC	Accuracy	-macro	Recall-macro	Precision-macro
MLR	0.9931 (0.0006)	0.9456 (0.0061)	0.9587 (0.0039)	0.9895 (0.0021)	0.9362 (0.0054)
MCSVM	0.9801 (0.0019)	0.9723 (0.0021)	0.9751 (0.0019)	0.9766 (0.0024)	0.9736 (0.0021)
XGBoost	0.9999 (0.0000)	0.9927 (0.0015)	0.9929 (0.0020)	0.9942 (0.0039)	0.9918 (0.0018)
LightGBM	0.9998 (0.0000)	0.9916 (0.0028)	0.9924 (0.0022)	0.9918 (0.0044)	0.9930 (0.0017)