Solving the class imbalance problem using ensemble algorithm: application of screening for aortic dissection

Table 7 Sensitivity (Se) and specificity (Sp) of ensemble learning models

	AdaBoost		EasyEnsemble		Ensemble model		Random forest
	Se	Sp	Se	Sp	Se	Sp	Se	Sp
1st	0.736	0.742	0.798	0.802	0.816	0.705	0.781	0.791
2nd	0.675	0.759	0.737	0.794	0.798	0.733	0.737	0.792
3rd	0.772	0.744	0.825	0.793	0.842	0.704	0.807	0.775
4th	0.631	0.765	0.702	0.816	0.807	0.724	0.693	0.789
5th	0.754	0.748	0.798	0.803	0.860	0.717	0.754	0.821
6th	0.631	0.762	0.781	0.802	0.825	0.730	0.728	0.810
7th	0.711	0.765	0.693	0.818	0.847	0.715	0.695	0.810
Average	70.1%	75.5%	76.1%	80.4%	82.8%	71.9%	74.2%	79.8%
Variance (\(\times {10}^{-3}\))	57.23	10.3	51.49	9.75	19.58	9.89	42.27	15.79

ISSN: 1472-6947