Informative gene selection and the direct classification of tumors based on relative simplicity

BMC Bioinformatics

Table 10 Test accuracy of different classifiers with informative genes selected by different feature-selection methods

Classifier	Feature-selection method	Leuk1	Lung1	Leuk2	SRBCT	Breast	Lung2	DLBCL	Cancers	GCM	Average
NB	ALL^a	85.29	81.25	100	60.00	66.67	88.06	86.67	79.73	52.17	77.76
	RS	94.12	84.38	100	85.00	93.33	88.06	90.00	85.14	71.74	87.97
	mRMR	79.41	68.75	100	90.00	93.33	97.01	96.67	70.27	45.65	82.34
	SVM-RFE	67.65	81.25	80.00	95.00	80.00	89.55	90.00	77.03	63.04	80.39
	TSG	91.18	84.38	93.33	100	86.67	94.03	100	71.62	65.22	87.38
	HC-K-TSP	91.18	81.25	100	80.00	80.00	95.52	86.67	77.03	65.22	84.10
KNN	ALL^a	67.65	75.00	86.67	70.00^b	63.33	88.06	93.33	64.86	34.78	71.71
	RS	97.06	78.13	93.33	90.00	93.33	95.52	93.33	72.97	43.48	84.13
	mRMR	70.59	68.75	80.00	80.00	96.67	86.57	100	54.05	36.96	74.84
	SVM-RFE	76.47	68.75	86.67	100	90.00	86.57	90.00	58.11	45.65	78.02
	TSG	91.18	75.00	93.33	100	80.00	88.06	96.67	74.32	39.13	81.97
	HC-K-TSP	88.24	87.50	86.67	85.00	83.33	94.03	93.33	64.86	52.17	81.68
SVM	ALL^a	79.41	87.50	100	100	83.33	97.01	100	83.78	65.22	88.47
	RS	94.12	84.38	100	95.00	93.33	95.52	96.67	89.19	65.22	90.38
	mRMR	76.47	78.13	100	75.00	96.67	95.52	96.67	71.62	45.65	81.75
	SVM-RFE	85.29	78.13	93.33	95.00	90.00	88.06	90.00	93.24	63.04	86.23
	TSG	91.18	81.25	93.33	80.00	80.00	94.03	100	68.92	54.35	82.56
	HC-K-TSP	85.29	84.38	100	90.00	86.67	98.51	96.67	82.43	60.87	87.20

^aResults reported in [6], ^bThe 30 reported in [3] is 70.00 after validation. Bold values indicate the best average accuracy in each classifier

ISSN: 1471-2105