Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 35 Results for the logistic regression classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0061	0.9256	0.0597	0.0744	0.9403	0.0511	0.9328	0.1503	0.0263
F-meas NC	0.2694	0.8297	0.0007	0.1703	0.9993	0.7427	0.9103	0.7474	0.6767
G-mean	0.0057	0.9286	0.0666	0.0714	0.9334	0.0463	0.9309	0.1425	0.0237
G-mean NC	0.0130	0.9012	0.0232	0.0988	0.9768	0.1257	0.9381	0.2417	0.0680
MCC	0.0056	0.9290	0.0668	0.0710	0.9332	0.0462	0.9310	0.1424	0.0237
MCC NC	0.0151	0.8969	0.0192	0.1031	0.9808	0.1488	0.9378	0.2646	0.0818
Precision	0.0061	0.9256	0.0597	0.0744	0.9403	0.0511	0.9328	0.1503	0.0263
Precision NC	1.0000	0.1928	0.0000	0.8072	1.0000	0.3080	0.4288	0.4105	0.9213
C	0.0122	0.9030	0.0235	0.0970	0.9765	0.1174	0.9389	0.2345	0.0628
D	0.5000	0.8069	0.0004	0.1931	0.9996	0.7859	0.8978	0.7862	0.7676