An Efficient Hybrid Feature Selection Method Using the Artificial Immune Algorithm for High-Dimensional Data

Table 3

Quantitative comparison between the optimal feature subset of HFIA and full features.

Id	Dataset	Features	avgNfs/std	ID	Dataset	Features	avgNfs/std
1	Leukemia1	5327	1.8/0.41	14	ALL-AML-3C	7129	1.8/0.41
2	DLBCL	5469	2.15/2.11	15	ALL-AML-4C	7129	3.6/1.9
3	9Tumor	5726	17.95/4.64	16	Lymphoma	4026	6.95/3.34
4	Brain Tumor1	5920	12.75/3.12	17	MLL	12582	1.8/0.41
5	Prostate tumor	10509	2.5/1.15	18	Ovarian	15154	1.65/0.49
6	Leukemia2	7129	7.4/5.34	19	SRBCT	2308	2.8/0.7
7	Brain Tumor2	10367	3.6/1.9	20	GLI-85	22283	2/0.67
8	Leukemia3	11225	1.95/0.6	21	NCI9	9712	15/6.19
9	11Tumor	12533	24.35/3.1	22	SMK-CAN-187	19993	6.2/2.94
10	Lung	12600	7.35/4.85	23	TOX-171	5748	19.6/7.87
11	Colon	2000	1.7/0.82	24	Orlraws10P	10304	3.7/0.95
12	CNS	7129	1/0	25	Pixraw10P	10000	2/0
13	ALL-AML	7129	1.1/0.32