Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 27 Results for the logistic regression classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.0496	0.9260	0.0560	0.0740	0.9440	0.0545	0.9349	0.1557	0.0281
F-meas NC	0.4944	0.8598	0.0036	0.1402	0.9964	0.4652	0.9254	0.5227	0.3275
G-mean	0.0441	0.9297	0.0642	0.0703	0.9358	0.0481	0.9326	0.1456	0.0247
G-mean NC	0.1006	0.9081	0.0277	0.0919	0.9723	0.1140	0.9395	0.2285	0.0613
MCC	0.0440	0.9297	0.0643	0.0703	0.9357	0.0480	0.9326	0.1455	0.0247
MCC NC	0.1212	0.9024	0.0218	0.0976	0.9782	0.1382	0.9393	0.2539	0.0756
Precision	0.0497	0.9258	0.0559	0.0742	0.9441	0.0545	0.9348	0.1558	0.0281
Precision NC	0.9541	0.5959	0.0005	0.4041	0.9995	0.6020	0.7487	0.6345	0.7548
C	0.1000	0.9081	0.0243	0.0919	0.9757	0.1153	0.9412	0.2326	0.0616
D	0.5000	0.8596	0.0033	0.1404	0.9967	0.4687	0.9254	0.5254	0.3262