A Novel Hierarchical Clustering Approach Based on Universal Gravitation

Table 3

Performance comparison on the synthetic datasets.

Dataset	K-means	K-means++	SC	DBSCAN	Birch	LGC	GHC
Figure 4(a)	0.8	0.8	0.99	0.99	1	1	1
	0.89	0.89	0.99	1	1	1	1
	0.81	0.81	0.99	0.99	1	1	1
	0.55	0.55	0.95	0.98	1	1	1
Figure 4(b)	0.92	0.92	1	0.99	0.97	0.99	1
	0.93	0.93	1	0.99	0.96	0.99	1
	0.79	0.79	0.99	0.97	0.92	0.97	0.99
	0.86	0.86	0.99	0.96	0.92	0.97	0.99
Figure 4(c)	0.89	0.89	1	1	0.91	1	1
	0.85	0.85	1	1	0.9	1	1
	0.81	0.81	1	1	0.85	1	1
	0.76	0.76	1	1	0.81	1	1
Figure 4(d)	0.77	0.76	0.99	0.99	0.84	1	1
	0.74	0.73	0.99	1	0.83	1	1
	0.53	0.53	0.99	0.99	0.69	1	1
	0.61	0.6	0.98	0.97	0.76	1	1
Figure 4(e)	1	1	1	1	1	1	1
	1	1	1	1	1	1	1
	1	1	1	1	1	1	1
	1	1	1	1	1	1	1
Figure 4(f)	0.86	0.86	0.99	1	1	1	1
	0.9	0.9	0.99	1	1	1	1
	0.81	0.81	0.99	1	1	1	1
	0.72	0.73	0.99	1	1	1	1
Figure 4(g)	0.96	0.96	1	1	1	1	1
	0.96	0.96	1	1	1	1	1
	0.92	0.92	1	1	1	1	1
	0.9	0.9	1	1	1	1	1
Figure 4(h)	0.5	0.5	1	1	0.51	1	1
	0.5	0.5	1	1	0.58	1	1
	0.5	0.5	1	1	0.54	1	1
	0	0	1	1	0.02	1	1
Figure 4(i)	0.75	0.75	1	1	0.86	1	1
	0.85	0.85	1	1	0.93	1	1
	0.75	0.75	1	1	0.86	1	1
	0.39	0.39	1	1	0.69	1	1
Figure 4(j)	0.5	0.5	0.77	0.59	0.5	0.53	1
	0.52	0.52	0.87	0.72	0.53	0.64	1
	0.52	0.52	0.79	0.63	0.56	0.48	1
	0	0	0.52	0.21	0	0.13	1

For each dataset, the numbers in the first row are RI, those in the second row are Purity, those in the third row are Fmeasure, and those in the forth row are NMI.