BMC Medical Informatics and Decision Making

Table 5 Performance comparison of five machine learning models with data balancing technology

From: Risk prediction of hyperuricemia based on particle swarm fusion machine learning solely dependent on routine blood tests

Models	Auc	Acc	P	R	F1
LR + smote	0.855	0.777	0.743	0.846	0.791
LR + adasyn	0.844	0.742	0.678	0.922	0.781
SVM + smote	0.972	0.921	0.921	0.921	0.921
SVM + adasyn	0.975	0.927	0.914	0.943	0.928
RF + smote	0.984	0.934	0.927	0.941	0.935
RF + adasyn	0.982	0.933	0.929	0.937	0.932
DNN + smote	0.958	0.958	0.947	0.969	0.954
DNN + adasyn	0.938	0.938	0.917	0.963	0.960
XGBoost + smote	0.997	0.973	0.975	0.971	0.973
XGBoost + adasyn	0.995	0.972	0.986	0.957	0.971

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com