Credit Risk Management of Consumer Finance Based on Big Data

Table 6

Comparison of the average performance evaluation index of each algorithm (rho = 0.5).

Rho = 0.5	Imbalance ratio	Accuracy	Recall rate	Precision	F1 measurement
SVM	10 : 1	0.8903	0.5234	0.5654	0.1211
Random forest		0.8852	0.4573	0.3891	0.3579
AdaBoost		0.9276	0.4978	0.7677	0.5537
Improve AdaBoost		0.8769	0.7851	0.5318	0.4891
SVM	20 : 1	0.9691	0.0785	1	0.1179
Random forest		0.9452	0.2967	0.5255	0.4123
AdaBoost		0.9783	0.4015	0.7976	0.6053
Improve AdaBoost		0.9216	0.6762	0.4013	0.5457