Table 2 The prediction performance of each model.

Model	AUC	Accuracy	Sensitivity	Specificity	Precision	F1 score	Brier score	C index
Train set
LR	0.784	0.727	0.828	0.625	0.688	0.752	0.187	0.784
SVM	0.782	0.766	0.797	0.734	0.750	0.773	0.186	0.782
GBM	0.865	0.789	0.844	0.734	0.761	0.800	0.146	0.865
NeuralNetwork	0.851	0.789	0.844	0.734	0.761	0.800	0.156	0.851
RandomForest	0.789	0.789	0.844	0.734	0.761	0.800	0.211	0.789
Xgboost	0.816	0.750	0.875	0.625	0.700	0.778	0.171	0.816
KNN	0.750	0.742	0.531	0.953	0.919	0.673	0.304	0.750
Adaboost	0.863	0.789	0.844	0.734	0.761	0.800	0.155	0.863
LightGBM	0.853	0.773	0.812	0.734	0.754	0.782	0.155	0.853
CatBoost	0.843	0.766	0.797	0.734	0.750	0.773	0.264	0.843
Test set
LR	0.661	0.679	0.643	0.714	0.692	0.667	0.238	0.661
SVM	0.773	0.750	0.786	0.714	0.733	0.759	0.196	0.773
GBM	0.778	0.786	0.786	0.786	0.786	0.786	0.184	0.778
NeuralNetwork	0.747	0.750	0.714	0.786	0.769	0.741	0.199	0.747
RandomForest	0.786	0.786	0.786	0.786	0.786	0.786	0.214	0.786
Xgboost	0.732	0.750	0.786	0.714	0.733	0.759	0.204	0.732
KNN	0.633	0.679	0.357	1. 000	1.000	0.526	0.410	0.633
Adaboost	0.763	0.786	0.786	0.786	0.786	0.786	0.197	0.763
LightGBM	0.778	0.786	0.786	0.786	0.786	0.786	0.180	0.778
CatBoost	0.778	0.786	0.786	0.786	0.786	0.786	0.264	0.778

LR, logistic regression; SVM, support vector machine; GBM, gradient boosting machine; Xgboost, eXtreme gradient boosting; KNN, k-nearest neighbors; AdaBoost, adaptive boosting; CatBoost, categorical boosting.

Quick links

Search