Table 5. Runtime cost (seconds) on real cancer gene expression dataset.

Dataset	HC	k-means	SOM	LAC	WSPA	RDCFCE	EM-PCE
BreastB	0.03	0.23	12.57	0.50	101.80	2510.98	302.70
DLBCLA	0.09	0.24	4.96	0.58	117.63	1946.50	236.34
Leukemia	0.33	0.90	13.10	3.28	412.93	4722.94	1144.20
NovartisBPLC	0.07	0.19	8.10	0.61	87.44	2172.79	354.17
Pomeroy2002v2	0.03	0.23	21.18	0.36	40.01	1381.63	217.04
Ramaswamy2001	0.28	2.21	30.93	5.51	631.94	5248.20	2747.36
Risinger2003	0.04	0.23	70.00	0.43	43.56	1727.51	228.74
Su2001	0.27	1.73	36.12	5.16	466.25	5491.07	2098.11
Overall	1.14	5.98	196.97	16.44	1910.56	10331.31	7328.66

The runtime costs of HC, k-means, SOM, LAC, WSPA, RDCFCE and EM-PCE on eight real gene expression datasets. RDCFCE costs more time than other methods.