BMC Medical Informatics and Decision Making

Table 2 The AUROC values of the base classifier algorithms and the ELAS

From: An ensemble learning with active sampling to predict the prognosis of postoperative non-small cell lung cancer patients

Task	Base classifier algorithms						ELAS
	SVM		L2-LR		CART		SVM-ELAS		L2-LR-ELAS		CART-ELAS
	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
1-year recurrence	0.649	0.063	0.660	0.072	0.603	0.072	0.702	0.079	0.674	0.071	0.668	0.056
1-year death	0.653	0.057	0.754	0.043	0.65	0.072	0.760	0.042	0.740	0.057	0.740	0.059
3-year recurrence	0.713	0.041	0.697	0.027	0.637	0.031	0.728	0.033	0.709	0.029	0.706	0.036
3-year death	0.702	0.044	0.711	0.040	0.663	0.041	0.733	0.035	0.720	0.037	0.737	0.040
5-year recurrence	0.751	0.053	0.730	0.061	0.668	0.045	0.748	0.055	0.735	0.063	0.724	0.051
5-year death	0.739	0.033	0.718	0.028	0.631	0.044	0.742	0.029	0.729	0.026	0.694	0.040
All tasks	0.701	0.063	0.711	0.056	0.642	0.057	0.736	0.052	0.718	0.055	0.711	0.054

The bold means the best results for corresponding tasks

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com