Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 19 Results for the logistic regression classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.3001	0.9321	0.0700	0.0679	0.9300	0.0452	0.9310	0.1405	0.0232
F-meas NC	0.4586	0.9189	0.0426	0.0811	0.9574	0.0774	0.9378	0.1857	0.0406
G-mean	0.2742	0.9356	0.0773	0.0644	0.9227	0.0410	0.9290	0.1334	0.0209
G-mean NC	0.4675	0.9185	0.0411	0.0815	0.9589	0.0792	0.9384	0.1883	0.0416
MCC	0.2685	0.9366	0.0792	0.0634	0.9208	0.0400	0.9286	0.1318	0.0205
MCC NC	0.5330	0.9116	0.0342	0.0884	0.9658	0.0944	0.9381	0.2062	0.0501
Precision	0.3101	0.9303	0.0667	0.0697	0.9333	0.0467	0.9317	0.1432	0.0240
Precision NC	0.8922	0.8272	0.0091	0.1728	0.9909	0.2769	0.8990	0.3673	0.1878
C	0.5000	0.9155	0.0350	0.0845	0.9650	0.0844	0.9398	0.1966	0.0443
D	0.5000	0.9155	0.0350	0.0845	0.9650	0.0844	0.9398	0.1966	0.0443