Causal relationship between systemic lupus erythematosus and primary liver cirrhosis based on two-sample bidirectional Mendelian randomization and transcriptome overlap analysis

Table 3 Models for identifying SLE patients using 9 hub genes based on 11 machine learning algorithms

Models	AUC	Accuracy	Sensitivity	Specificity	Recall	F1	Cohorts
LR	0.927	0.906	0.907	0.895	0.907	0.947	T
LR	0.966	0.910	0.908	0.933	0.908	0.949	V
NaiveBayes	0.921	0.870	0.868	0.911	0.868	0.925	T
NaiveBayes	0.966	0.879	0.870	1.000	0.870	0.931	V
SVM	0.958	0.955	0.957	0.930	0.957	0.975	T
SVM	0.940	0.965	0.973	0.867	0.973	0.981	V
KNN	0.982	0.891	0.882	1.000	0.882	0.938	T
KNN	0.976	0.879	0.870	1.000	0.870	0.930	V
RandomForest	1.000	0.995	0.995	1.000	0.995	0.997	T
RandomForest	0.938	0.940	0.940	1.000	0.940	0.966	V
ExtraTrees	1.000	1.000	1.000	1.000	1.000	1.000	T
ExtraTrees	0.947	0.950	0.951	1.000	0.951	0.972	V
XGBoost	0.996	0.995	0.996	0.982	0.996	0.997	T
XGBoost	0.964	0.920	0.918	0.933	0.918	0.955	V
LightGBM	0.993	0.967	0.968	0.965	0.968	0.982	T
LightGBM	0.967	0.925	0.924	0.933	0.924	0.958	V
GradientBoosting	0.979	0.951	0.951	0.947	0.951	0.973	T
GradientBoosting	0.934	0.915	0.913	1.000	0.913	0.952	V
AdaBoost	0.972	0.905	0.900	0.965	0.900	0.946	T
AdaBoost	0.974	0.930	0.924	1.000	0.924	0.960	V
MLP	0.938	0.897	0.895	0.930	0.895	0.942	T
MLP	0.961	0.925	0.924	0.933	0.924	0.958	V

ISSN: 1478-6362