A Novel Convex Clustering Method for High-Dimensional Data Using Semiproximal ADMM

Table 3

Results for Case III. Empirical mean and standard deviation (SD) of the RI, FMI, FNR, and FPR based on 50 repetitions.

Case III:
	Methods	RI		FMI		FNR		FPR
		Mean	SD	Mean	SD	Mean	SD	Mean	SD
2000	K-means	0.903	0.081	0.837	0.134	0.000	0.000	1.000	0.000
	CC	0.966	0.063	0.950	0.085	0.000	0.000	1.000	0.000
	SCC	0.988	0.027	0.977	0.045	0.015	0.075	0.506	0.115
	SGLCC	0.993	0.006	0.987	0.012	0.029	0.115	0.128	0.070
3000	K-means	0.914	0.085	0.860	0.136	0.000	0.000	1.000	0.000
	CC	0.947	0.071	0.924	0.096	0.000	0.000	1.000	0.000
	SCC	0.989	0.015	0.978	0.031	0.007	0.035	0.775	0.125
	SGLCC	0.993	0.019	0.986	0.030	0.029	0.114	0.089	0.041
4000	K-means	0.897	0.080	0.834	0.122	0.000	0.000	1.000	0.000
	CC	0.915	0.097	0.793	0.253	0.000	0.000	1.000	0.000
	SCC	0.933	0.080	0.825	0.228	0.025	0.101	0.823	0.231
	SGLCC	0.983	0.060	0.979	0.070	0.030	0.119	0.077	0.028
5000	K-means	0.892	0.085	0.825	0.136	0.000	0.000	1.000	0.000
	CC	0.900	0.048	0.764	0.115	0.000	0.000	1.000	0.000
	SCC	0.938	0.058	0.856	0.137	0.000	0.000	0.780	0.214
	SGLCC	0.999	0.003	0.999	0.006	0.010	0.068	0.003	0.001