Effective hybrid feature selection using different bootstrap enhances cancers classification performance

Abdelwahed, Noura Mohammed; El-Tawel, Gh. S.; Makhlouf, M. A.

doi:10.1186/s13040-022-00304-y

BioData Mining

Table 6 Average results after applying O/IFBS-RFS-RFE after 20 runs

From: Effective hybrid feature selection using different bootstrap enhances cancers classification performance

Algo.	Train Data %	Test Data %	Over-fitting Diff. %	Pre	Rec	F1-score	NO.F	F-Time (sec)	C-Time (sec)	AUC	Var.	ACC %
RNA gene dataset
LR Classifier
O/IFBS-RFS-	100.000	99.975	0.025	0.999	0.999	0.999	238.800	4.220	0.176	1.000	0.0000006	99.975
O/IFBS-RFS-RFE	100.000	99.994	0.006	0.999	0.999	0.999	119.200	13.726	0.307	1.000	0.0000004	99.994
SVM Classifier
O/IFBS-RFS-	100.000	99.950	0.05	0.999	0.999	0.999	238.800	4.220	0.197	1.000	0.0000025	99.950
O/IFBS-RFS-RFE	100.000	99.981	0.019	0.999	0.999	0.999	119.200	13.726	0.125	1.000	0.0000004	99.981
RF Classifier
O/IFBS-RFS-	100.000	99.888	0.112	0.999	0.999	0.999	238.800	4.220	0.755	1.000	0.0000076	99.888
O/IFBS-RFS-RFE	100.000	99.913	0.087	0.999	0.999	0.999	119.200	13.726	0.596	0.999	0.0000054	99.913
Bagg Classifier
O/IFBS-RFS-	99.974	99.357	0.617	0.994	0.992	0.993	238.800	4.220	0.513	0.999	0.0000828	99.357
O/IFBS-RFS-RFE	99.972	99.363	0.609	0.994	0.993	0.993	119.200	13.726	0.266	0.999	0.000083	99.363
DNA CNV dataset
LR Classifier
O/IFBS-RFS-	92.581	89.818	2.763	0.904	0.861	0.877	973.000	3.650	3.850	0.975	0.00031	89.818
O/IFBS-RFS-RFE	91.878	89.601	2.277	0.906	0.857	0.885	485.000	1460	1.950	0.936	0.00035	89.601
SVM Classifier
O/IFBS-RFS-	93.361	90.253	3.108	0.917	0.860	0.878	973.000	3.650	22.00	0.980	0.00065	90.253
O/IFBS-RFS-RFE	94.241	90.979	3.262	0.925	0.873	0.891	485.000	1460	11.700	0.985	0.00028	90.979
RF Classifier
O/IFBS-RFS-	95.527	90.764	4.763	0.914	0.868	0.882	973.000	3.650	2.650	0.984	0.00027	90.764
O/IFBS-RFS-RFE	95.681	90.954	4.727	0.919	0.872	0.890	485.000	1460	1.750	0.941	0.00027	90.954
Bagg Classifier
O/IFBS-RFS-	97.958	92.712	5.246	0.926	0.906	0.913	973.000	3.650	6.550	0.980	0.00027	92.712
O/IFBS-RFS-RFE	95.318	92.834	2.484	0.927	0.906	0.913	485.000	1460	3.150	0.980	0.00027	92.834
Parkinson’s disease dataset
LR classifier
O/IFBS-RFS-	79.050	78.482	0.568	0.742	0.619	0.626	155.50	1.058	0.093	0.764	0.00123	78.482
O/IFBS-RFS-RFE	77.744	77.427	0.317	0.712	0.597	0.598	77.550	5.551	0.118	0.731	0.00092	77.427
SVM classifier
O/IFBS-RFS-	76.009	75.442	0.567	0.612	0.539	0.508	155.500	1.058	0.511	0.637	0.00041	75.442
O/IFBS-RFS-RFE	77.500	76.672	0.828	0.653	0.566	0.542	77.550	5.551	0.420	0.669	0.00051	76.672
RF classifier
O/IFBS-RFS-	100.000	94.494	5.506	0.945	0.909	0.924	155.500	1.058	1.122	0.985	0.00064	94.494
O/IFBS-RFS-RFE	100.000	94.082	5.918	0.943	0.901	0.917	77.550	5.551	0.911	0.983	0.00070	94.082
Bagg Classifier
O/IFBS-RFS-	99.720	93.196	6.524	0.916	0.906	0.909	155.500	1.058	1.091	0.965	0.00093	93.196
O/IFBS-RFS-RFE	99.719	92.917	6.802	0.914	0.900	0.905	77.550	5.550	0.511	0.966	0.00084	92.917
Dermatology erythemato-squamous diseases dataset
LR classifier
O/IFBS-RFS-	96.691	96.441	0.250	0.649	0.624	0.630	11.000	0.167	0.025	0.998	0.000848	96.441
O/IFBS-RFS-RFE	92.532	92.350	0.212	0.801	0.751	0.766	10.000	0.500	0.128	0.999	0.000790	92.350
SVM classifier
O/IFBS-RFS-	95.082	95.000	0.082	0.638	0.608	0.613	11.000	0.167	0.025	0.977	0.000632	95.000
O/IFBS-RFS-RFE	98.361	98.356	0.005	0.892	0.900	0.895	10.000	0.500	0.047	0.999	0.001040	98.356
RF classifier
O/IFBS-RFS-	100.000	100.000	0.0	1.000	1.000	1.000	11.000	0.167	0.562	1.000	0.0	100.00
O/IFBS-RFS-RFE	100.000	100.000	0.0	1.000	1.000	1.000	10.000	0.500	0.500	1.000	0.0	100.00
Bagg classifier
O/IFBS-RFS-	100.000	100.000	0.0	1.000	1.000	1.000	11.000	0.167	0.520	0.999	0.0	100.000
O/IFBS-RFS-RFE	100.000	100.000	0.0	1.000	1.000	1.000	10.000	0.500	0.500	0.991	0.0	100.000
BreastEw dataset
LR classifier
O/IFBS-RFS-	94.647	94.148	0.499	0.944	0.932	0.936	22.900	0.399	0.010	0.988	0.00095	94.148
O/IFBS-RFS-RFE	95.305	94.842	0.463	0.949	0.942	0.944	11.300	0.1033	0.091	0.992	0.00086	94.842
SVM classifier
O/IFBS-RFS-	92.110	91.889	0.221	0.934	0.897	0.909	22.900	0.399	0.067	0.978	0.00098	91.889
O/IFBS-RFS-RFE	93.515	93.400	0.115	0.943	0.918	0.927	11.300	0.103	0.058	0.983	0.00094	93.400
RF classifier
O/IFBS-RFS-	99.563	97.500	2.063	0.981	0.976	0.977	22.900	0.3994	0.411	0.996	0.00031	97.500
O/IFBS-RFS-RFE	100.000	98.000	2.000	0.979	0.977	0.978	11.300	0.103	0.404	0.997	0.00031	98.000
Bagg Classifier
O/IFBS-RFS-	99.819	97.618	2.201	0.977	0.973	0.974	22.900	0.399	0.089	0.994	0.00038	97.618
O/IFBS-RFS-RFE	99.803	97.505	2.298	0.976	0.972	0.973	11.300	0.103	0.065	0.993	0.00034	97.505

Back to article page

ISSN: 1756-0381

Contact us

General enquiries: journalsubmissions@springernature.com