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 9 The proposed methods compared with the MIFS method

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

Datasets	Train Data %	Test Data %	Over-fitting Diff. %	Pre	Rec	F1-score	NO.F	F-Time (sec)	C-Time (sec)	AUC	Var.	ACC %
LR classifier
RNA gene	100.000	99.875	0.125	0.999	0.998	0.988	10,000	192.552	2.896	1.000	0.000016	99.875
DNA CNV	96.597	84.978	11.619	0.817	0.782	0.788	9000	173.955	25.195	0.954	0.000416	84.978
Parkinson’s disease	77.058	75.525	1.533	0.620	0.556	0.538	300	0.377	0.037	0.682	0.001001	75.525
Dermatology diseases	97.845	96.989	0.856	0.971	0.965	0.966	25	0.203	0.003	0.997	0.000585	96.989
BreastEW	94.396	93.678	0.718	0.938	0.928	0.932	20	0.067	0.002	0.988	0.000694	93.678
SVM classifier
RNA gene	100.000	99.750	0.250	0.998	0.997	0.997	10,000	192.552	2.534	1.000	0.000028	99.750
DNA CNV	91.606	84.122	7.484	0.860	0.756	0.775	9000	173.955	75.394	0.949	0.000668	84.122
Parkinson’s disease	75.676	72.228	3.448	0.472	0.498	0.448	300	0.203	0.138	0.627	0.000814	72.228
Dermatology diseases	98.421	97.523	0.898	0.976	0.967	0.969	25	0.203	0.028	0.998	0.000924	97.523
BreastEW	92.013	91.563	0.450	0.929	0.895	0.906	20	0.067	0.017	0.976	0.001014	91.563
RF classifier
RNA gene	100.000	99.627	0.373	0.998	0.996	0.997	10,000	192.552	1.252	1.000	0.000036	99.627
DNA CNV	92.962	80.623	12.339	0.771	0.719	0.718	9000	173.955	3.528	0.942	0.000614	80.623
Parkinson’s disease	100.000	84.782	15.218	0.827	0.748	0.773	300	0.377	0.376	0.876	0.002303	84.782
Dermatology diseases	100.000	96.456	3.544	0.972	0.950	0.955	25	0.203	0.148	0.999	0.001473	96.456
BreastEW	100.000	96.140	3.860	0.963	0.956	0.958	20	0.067	0.110	0.990	0.000944	96.140
Bagg classifier
RNA gene	99.847	98.628	1.219	0.989	0.985	0.987	10,000	192.552	7.322	0.999	0.000036	98.628
DNA CNV	98.960	78.806	20.154	0.733	0.699	0.707	9000	173.955	25.309	0.912	0.000613	78.806
Parkinson’s disease	99.574	79.239	20.335	0.729	0.729	0.727	300	0.377	0.673	0.794	0.002002	79.239
Dermatology diseases	99.696	95.105	4.591	0.955	0.940	0.939	25	0.203	0.021	0.995	0.001473	95.105
BreastEW	99.492	95.435	4.057	0.957	0.947	0.950	20	0.067	0.022	0.986	.000901	95.435

Back to article page

ISSN: 1756-0381

Contact us

General enquiries: journalsubmissions@springernature.com