Threshold optimization and random undersampling for imbalanced credit card data

Journal of Big Data

Table 24 Results for the random forest depth 4 classifier

Technique	Threshold	TPR	FPR	FNR	TNR	F-meas	G-mean	MCC	Precision
F-meas	0.1161	0.9254	0.0728	0.0746	0.9272	0.0427	0.9261	0.1358	0.0219
F-meas NC	0.2347	0.8850	0.0183	0.1150	0.9817	0.1550	0.9319	0.2688	0.0857
G-mean	0.1073	0.9305	0.0853	0.0695	0.9147	0.0371	0.9224	0.1258	0.0189
G-mean NC	0.1551	0.9069	0.0421	0.0931	0.9579	0.0755	0.9318	0.1822	0.0396
MCC	0.1072	0.9305	0.0854	0.0695	0.9146	0.0370	0.9223	0.1256	0.0189
MCC NC	0.1607	0.9045	0.0393	0.0955	0.9607	0.0809	0.9319	0.1889	0.0426
Precision	0.1162	0.9254	0.0727	0.0746	0.9273	0.0427	0.9262	0.1359	0.0219
Precision NC	0.4851	0.8484	0.0038	0.1516	0.9962	0.5213	0.9190	0.5716	0.4317
C	0.2500	0.8811	0.0149	0.1189	0.9851	0.1724	0.9315	0.2867	0.0958
D	0.5000	0.8490	0.0015	0.1510	0.9985	0.6337	0.9206	0.6554	0.5103