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

Table 7 The paired student t-test results between the benchmark algorithms and the ELAS

Metric	Comparison	1-year tasks	3-year tasks	5-year tasks	All tasks
AUROC	SVM-AdaBoost versus SVM-ELAS	0.231	< 0.01	0.041	< 0.01
	SVM-Bagging versus SVM-ELAS	< 0.01	< 0.01	0.490	< 0.01
	SVM-SMOTE versus SVM-ELAS	< 0.01	< 0.01	0.104	< 0.01
	SVM-TomekLinks versus SVM-ELAS	< 0.01	< 0.01	0.454	< 0.01
AUPRC	SVM-AdaBoost versus SVM-ELAS	0.428	< 0.01	0.096	< 0.01
	SVM-Bagging versus SVM-ELAS	0.396	0.046	0.146	0.337
	SVM-SMOTE versus SVM-ELAS	< 0.01	< 0.01	0.084	< 0.01
	SVM-TomekLinks versus SVM-ELAS	0.334	0.041	0.287	0.088

The bold means the p-value is less than 0.05, which means the results between different models have statistically significant differences

ISSN: 1472-6947