Table 3.

Model performances for frequent users 5.

Model	AUC	SEN	SPE	PPV	NPV
LR	81.3 (80.4–82.2)	67.7	81.3	10.1	98.8
GBM	81.4 (80.4–82.3)	69.9	78.9	9.3	98.8
NB	61.2 (60.4–62.0)	24.8	97.6	24.5	97.7
NN	78.4 (77.3–79.4)	63.2	82.1	9.9	98.6
RF 1	51.4 (51.1–51.8)	2.9	99.9	51.3	97.1
RF 2	81.1 (80.2–82.0)	72.5	75.6	8.4	98.9

GBM gradient boosting machine, LR logistic regression, NB naïve bayes, NN neural network, RF random forests (1: binary outcome, 2: continuous outcome). AUC area under the curve, SEN sensitivity, SPE specificity, PPV positive predicted value, NPV negative predicted value.