Default Risk Prediction of Enterprises Based on Convolutional Neural Network in the Age of Big Data: Analysis from the Viewpoint of Different Balance Ratios

Table 9

The performance of CNN and other four models on J₁₃.

Method	G-mean	Accuracy	Type-II error	AUC
CNN	0.775	0.727	0.143	0.779
KNN	0.663	0.785	0.476	0.682
DT	0.679	0.818	0.476	0.702
SVM	0.741	0.678	0.143	0.749
LR	0.735	0.719	0.238	0.736

The bold value means that the method ranks first in some evaluation criterion. For example, CNN ranks first in terms of G-mean, accuracy, and AUC, which indicates that its comprehensive ranking also ranks first.