CNFE-SE: a novel approach combining complex network-based feature engineering and stacked ensemble to predict the success of intrauterine insemination and ranking the features

Table 6 Results of the A-Test: The values of Γ_ζ^ and the minimum value of Γ_ζ

Feature set	Classifier	Γ_ζ^	Minimum of Γ_ζ
All 296 features	RF	26.1	12.8
	DT	24.6	14.1
	NB	24.9	12.6
	ANN	25.5	14.5
	SVM	24.7	20.7
	XGboost	24.3	15.4
	LGBM	24.6	14.2
	Adaboost	25.8	12.7
	CNFE-SE without FE	16.9	11.7
	CNFE-SE with FE	11.3	6.8
Only most important features	RF	24.7	13.5
	DT	24.3	14.9
	NB	23.6	16.1
	ANN	24.6	15.8
	SVM	23.7	16.4
	XGboost	23.1	15.9
	LGBM	22.3	13.7
	Adaboost	21.5	13.6
	CNFE-SE without FE	16.1	11.3
	CNFE-SE with FE	10.9	6.2

ISSN: 1472-6947