Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 31 Results for the logistic regression classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0170	0.9270	0.0587	0.0730	0.9413	0.0520	0.9340	0.1519	0.0268
F-meas NC	0.3684	0.8465	0.0013	0.1535	0.9987	0.6562	0.9193	0.6736	0.5395
G-mean	0.0156	0.9305	0.0650	0.0695	0.9350	0.0474	0.9326	0.1446	0.0243
G-mean NC	0.0355	0.9041	0.0252	0.0959	0.9748	0.1181	0.9386	0.2338	0.0635
MCC	0.0155	0.9307	0.0654	0.0693	0.9346	0.0472	0.9325	0.1443	0.0242
MCC NC	0.0422	0.8994	0.0201	0.1006	0.9799	0.1411	0.9386	0.2578	0.0769
Precision	0.0170	0.9270	0.0587	0.0730	0.9413	0.0520	0.9340	0.1519	0.0268
Precision NC	0.9873	0.3665	0.0001	0.6335	0.9999	0.4655	0.5745	0.5312	0.8709
C	0.0357	0.9037	0.0231	0.0963	0.9769	0.1193	0.9394	0.2367	0.0639
D	0.5000	0.8327	0.0009	0.1673	0.9991	0.7092	0.9119	0.7171	0.6197