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 5 Average results after applying 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
IFBS-RFS	100.000	99.925	0.075	0.999	0.999	0.999	239.000	5.421	0.193	1.000	0.000004	99.925
IFBS-RFS-RFE	100.000	99.975	0.025	0.999	0.999	0.999	125.250	15.201	0.357	1.000	0.0000004	99.975
SVM Classifier
IFBS-RFS	99.999	99.906	0.093	0.999	0.998	0.999	239.000	5.421	0.225	1.000	0.000005	99.906
IFBS-RFS-RFE	100.000	99.988	0.012	0.999	0.999	0.999	125.250	15.201	0.153	1.000	0.0000002	99.988
RF Classifier
IFBS-RFS	100.000	99.694	0.306	0.998	0.997	0.997	239.000	5.421	0.901	1.000	0.000030	99.694
IFBS-RFS-RFE	100.000	99.807	0.193	0.999	0.998	0.998	125.250	15.201	0.737	0.999	0.000019	99.807
Bagg Classifier
IFBS-RFS	99.947	99.002	0.945	0.991	0.989	0.989	239.000	5.421	0.635	0.999	0.000075	99.002
IFBS-RFS-RFE	99.955	99.027	0.928	0.992	0.989	0.990	125.250	15.201	0.327	0.999	0.000075	99.027
DNA CNV dataset
LR Classifier
IFBS-RFS	88.525	82.889	5.636	0.812	0.752	0.763	966.000	6.250	3.876	0.942	0.000370	82.889
IFBS-RFS-RFE	88.000	84.000	4.000	0.831	0.764	0.804	482.000	1545	2.149	0.959	0.000420	84.000
SVM Classifier
IFBS-RFS	88.341	81.637	6.704	0.815	0.729	0.745	966.000	6.050	29.078	0.955	0.000580	81.637
IFBS-RFS-RFE	89.668	82.268	7.400	0.827	0.738	0.753	482.000	1545	15.721	0.960	0.000880	82.268
RF Classifier
IFBS-RFS	89.660	80.089	9.571	0.768	0.7025	0.709	966.000	6.050	3.150	0.938	0.000410	80.089
IFBS-RFS-RFE	89.935	80.138	9.797	0.770	0.703	0.719	482.000	1545	2.487	0.941	0.000470	80.138
Bagg Classifier
IFBS-RFS	97.697	78.316	19.381	0.733	0.722	0.702	966.000	6.050	7.850	0.867	0.000450	78.316
IFBS-RFS-RFE	89.035	78.309	10.726	0.730	0.695	0.702	482.000	1545	4.023	0.910	0.000480	78.309
Parkinson’s disease dataset
LR Classifier
IFBS-RFS	78.122	76.718	1.404	0.664	0.588	0.582	154.263	1.071	0.082	0.732	0.002210	76.718
IFBS-RFS-RFE	76.693	73.998	2.695	0.667	0.588	0.581	80.050	4.594	0.164	0.722	0.001990	73.998
SVM Classifier
IFBS-RFS	75.684	72.248	3.436	0.468	0.498	0.448	154.263	1.071	0.482	0.621	0.000770	72.248
IFBS-RFS-RFE	75.697	72.228	3.469	0.464	0.497	0.448	80.050	4.594	0.407	0.619	0.000760	72.228
RF Classifier
IFBS-RFS	99.999	83.912	16.087	0.811	0.738	0.760	154.263	1.071	1.230	0.866	0.003480	83.912
IFBS-RFS-RFE	100.000	81.485	18.515	0.773	0.700	0.719	80.050	4.594	0.964	0.834	0.003500	81.485
Bagg Classifier
IFBS-RFS	99.590	80.810	17.78	0.754	0.731	0.737	154.263	1.071	1.225	0.826	0.003460	80.810
IFBS-RFS-RFE	99.580	79.191	20.389	0.729	0.707	0.713	80.050	4.594	0.546	0.804	0.003440	79.191
Dermatology erythemato-squamous diseases dataset
LR classifier
IFBS-RFS	91.531	91.000	0.531	0.771	0.796	0.777	13.000	0.515	0.002	0.988	0.000881	91.000
IFBS-RFS-RFE	92.198	91.801	0.397	0.773	0.799	0.780	12.000	0.016	0.020	0.988	0.000975	91.801
SVM classifier
IFBS-RFS	94.870	93.979	0.891	0.888	0.878	0.875	13.000	0.515	0.023	0.988	0.001285	93.979
IFBS-RFS-RFE	94.869	93.979	0.890	0.888	0.878	0.875	12.000	0.016	0.075	0.989	0.001285	93.979
RF classifier
IFBS-RFS	97.025	93.183	3.482	0.900	0.892	0.889	13.000	0.515	0.142	0.984	0.001493	93.183
IFBS-RFS-RFE	97.000	93.500	3.500	0.900	0.892	0.889	12.000	0.016	0.140	0.980	0.001490	93.500
Bagg classifier
IFBS-RFS	96.903	92.102	4.801	0.895	0.884	0.881	13.000	0.515	0.016	0.989	0.003297	92.102
IFBS-RFS-RFE	97.177	81.194	15.983	0.789	0.764	0.760	12.000	0.016	0.014	0.970	0.081251	81.194
BreastEW dataset
LR Classifier
IFBS-RFS	94.394	93.678	0.461	0.938	0.938	0.938	23.100	0.410	0.012	0.988	0.000690	93.678
IFBS-RFS-RFE	94.855	94.403	0.452	0.946	0.946	0.946	11.900	0.103	0.091	0.992	0.000520	94.403
SVM Classifier
IFBS-RFS	92.010	91.563	0.447	0.929	0.929	0.929	23.100	0.410	0.069	0.976	0.001010	91.563
IFBS-RFS-RFE	93.888	93.503	0.385	0.944	0.944	0.944	11.900	0.103	0.059	0.983	0.000550	93.503
RF Classifier
IFBS-RFS	100.000	96.411	3.589	0.965	0.965	0.965	23.100	0.410	0.452	0.991	0.000980	96.411
IFBS-RFS-RFE	100.000	95.277	4.723	0.952	0.952	0.952	11.900	0.103	0.433	0.989	0.000930	95.277
Bagg Classifier
IFBS-RFS	99.625	95.302	4.323	0.954	0.954	0.954	23.100	0.410	0.099	0.985	0.000920	95.302
IFBS-RFS-RFE	99.610	94.416	5.194	0.944	0.944	0.944	11.900	0.103	0.085	0.981	0.001170	94.416

Back to article page

ISSN: 1756-0381

Contact us

General enquiries: journalsubmissions@springernature.com