[Retracted] Big Data Analytics for Complex Credit Risk Assessment of Network Lending Based on SMOTE Algorithm

Table 3

Parameter selection and important variables of each model.

Model	Original training set		New training set
	Parameter selection	Important variables	Parameter selection	Important variables
CART	—	Succeed, field	—	Succeed
C4.5	—	Succeed, field, size	—	Autoloan, score
AdaBoost	—	Succeed, field, tile, application, size, score	—	Succeed, empLength, paid, size, grade, borrowType
SVM	C: classification mode, polynomial kernel function	—	C: classification mode, polynomial kernel function, weight is 2 : 1.4	—
ANN	Number of hidden nodes = 6, maximum iteration times = 200	—	Number of hidden nodes = 11, maximum iteration times = 207	—
RF	Number of spanning trees = 800, number of variables selected by node branches = 25	Paid, succeed, score, field, application, grade	Number of spanning trees = 800, number of variables selected by node branches = 3	Paid, succeed, application, score, size, grade