Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 23 Results for the logistic regression classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.1336	0.9307	0.0588	0.0693	0.9412	0.0524	0.9358	0.1528	0.0270
F-meas NC	0.4471	0.8918	0.0134	0.1082	0.9866	0.2030	0.9379	0.3143	0.1175
G-mean	0.1215	0.9337	0.0659	0.0663	0.9341	0.0475	0.9338	0.1447	0.0244
G-mean NC	0.2451	0.9130	0.0299	0.0870	0.9701	0.1027	0.9410	0.2172	0.0547
MCC	0.1210	0.9340	0.0663	0.0660	0.9337	0.0473	0.9337	0.1444	0.0243
MCC NC	0.2696	0.9087	0.0271	0.0913	0.9729	0.1141	0.9401	0.2292	0.0613
Precision	0.1337	0.9307	0.0588	0.0693	0.9412	0.0525	0.9358	0.1528	0.0270
Precision NC	0.9082	0.7439	0.0023	0.2561	0.9977	0.5093	0.8474	0.5555	0.4803
C	0.2500	0.9120	0.0272	0.0880	0.9728	0.1050	0.9418	0.2214	0.0558
D	0.5000	0.8880	0.0110	0.1120	0.9890	0.2218	0.9370	0.3324	0.1274