Research on Multilevel Classification of High-Speed Railway Signal Equipment Fault Based on Text Mining

Table 7

Experimental results of ensemble learning model classification.

Method	Level	Accuracy	Recall	F1 value
RF	First-level classification	0.8516	0.8419	0.8467
	Second-level classification	0.7526	0.7012	0.7568
GBDT	First-level classification	0.8598	0.8316	0.8455
	Second-level classification	0.7614	0.7278	0.7545
Stacking + multitask cooperative voting decision tree	First-level classification	0.8924	0.8561	0.8738
	Second-level classification	0.8152	0.7446	0.7783