[Retracted] The Application of the Data Mining Method Based on Ensemble Learning to the Research on Early Warning of Financial Crisis in Small- and Medium-Sized Enterprises

Table 3

Prediction results of different models based on fused multisource information and based on fused multisource information.

Train-test ratio	Accuracy	Model
		BPNN	SVM	KNN	LOG	MDA	Integrated learning
60 : 40	Acc-DF	0.8750	0.8125	0.6875	0.6250	0.8125	0.8750
	Acc-NF	0.8250	0.9250	0.9500	0.8250	0.9250	0.9250
	Acc	0.8393	0.8929	0.8750	0.7679	0.8929	0.9107
	-measure	0.7638	0.8125	0.7627	0.6063	0.8125	0.8489
70 : 30	Acc-DF	0.8333	0.9167	0.8333	0.5000	0.8333	0.9167
	Acc-NF	0.8667	0.9000	0.8667	0.8333	0.9333	0.9333
	Acc	0.8571	0.9048	0.8571	0.7381	0.9048	0.9286
	-measure	0.7715	0.8487	0.7715	0.5222	0.8333	0.8807