Symmetry Based Automatic Evolution of Clusters: A New Approach to Data Clustering

Table 2

(a) Average of adjusted Rand index for SBKM, MOCK, and MOLGC. (b) Standard deviation of adjusted Rand index for SBKM, MOCK, and MOLGC.

(a)

Data sets	Number of points	Number of dimensions	Number of clusters	Average value of adjusted Rand index
	()	()	()	SBKM	MOCK	MOLGC
Data set-1	200	2	2	0.7585	0.9840	0.9845
Data set-2	350	2	3	0.7491	0.9245	0.9515
Data set-3	250	2	5	0.5158	0.9510	0.9910
Data set-4	1000	2	4	0.8605	0.9815	0.9816
Data set-5	838	10	4	0.7225	0.9862	0.9895
Data set-6	3050	10	10	0.6585	0.9673	0.9795
Data set-7	351	50	4	0.6775	1.0000	1.0000
Data set-8	2328	50	10	0.6325	0.9950	0.9955
Iris	150	4	3	0.7685	0.9350	0.9810
Cancer	683	9	2	0.7877	0.9520	0.9740
Wine	178	13	3	0.6591	0.9575	0.9585
Diabetes	768	8	2	0.7111	0.9840	0.9910

(b)

Data sets	Number of points	Number of dimensions	Number of clusters	Standard deviation of adjusted Rand index
	()	()	()	SBKM	MOCK	MOLGC
Data set-1	200	2	2	0.075	0.040	0.035
Data set-2	350	2	3	0.081	0.055	0.045
Data set-3	250	2	5	0.090	0.078	0.050
Data set-4	1000	2	4	0.121	0.095	0.055
Data set-5	838	10	4	0.125	0.085	0.060
Data set-6	3050	10	10	0.150	0.070	0.028
Data set-7	351	50	4	0.175	0.075	0.041
Data set-8	2328	50	10	0.190	0.090	0.035
Iris	150	4	3	0.080	0.036	0.022
Cancer	683	9	2	0.090	0.045	0.037
Wine	178	13	3	0.085	0.051	0.035
Diabetes	768	8	2	0.070	0.050	0.030