A Novel Hierarchical Clustering Approach Based on Universal Gravitation
Table 4
Performances of different clustering approaches on UCI datasets tested.
Dataset
K-means
K-means++
SC
DBSCAN
Birch
LGC
GHC
BTissue
0.71
0.6
0.65
0.66
0.7
0.73
0.78
0.43
0.39
0.34
0.42
0.52
0.5
0.57
0.36
0.31
0.25
0.44
0.44
0.39
0.47
0.36
0.29
0.14
0.46
0.53
0.43
0.54
Ecoli
0.81
0.81
0.86
0.77
0.9
0.88
0.91
0.82
0.83
0.79
0.62
0.8
0.82
0.86
0.54
0.53
0.71
0.64
0.83
0.79
0.82
0.61
0.61
0.6
0.47
0.73
0.64
0.7
Iris
0.88
0.88
0.88
0.77
0.82
0.89
0.97
0.89
0.89
0.89
0.68
0.86
0.93
0.97
0.82
0.82
0.82
0.73
0.72
0.82
0.95
0.76
0.76
0.75
0.66
0.67
0.75
0.9
Wine
0.72
0.72
0.56
0.34
0.93
0.62
0.95
0.7
0.7
0.53
0.4
0.96
0.66
0.97
0.58
0.58
0.45
0.51
0.9
0.58
0.93
0.43
0.43
0.12
0
0.82
0.42
0.88
PID
0.55
0.55
0.54
0.55
0.57
0.56
0.6
0.66
0.66
0.65
0.65
0.7
0.68
0.73
0.63
0.63
0.7
0.71
0.49
0.58
0.66
0.03
0.03
0
0
0.08
0.05
0.11
BSTC
0.6
0.6
0.63
0.63
0.61
0.64
0.64
0.76
0.76
0.76
0.76
0.77
0.76
0.77
0.71
0.71
0.77
0.76
0.7
0.78
0.78
0.02
0.02
0
0.01
0.05
0
0.02
Glass
0.68
0.68
0.71
0.63
0.71
0.58
0.72
0.59
0.59
0.59
0.49
0.6
0.68
0.64
0.5
0.5
0.41
0.43
0.48
0.49
0.45
0.42
0.42
0.37
0.32
0.41
0.34
0.4
Cancer
0.92
0.92
0.93
0.89
0.87
0.95
0.94
0.96
0.96
0.96
0.94
0.96
0.97
0.97
0.93
0.93
0.94
0.9
0.87
0.95
0.94
0.74
0.73
0.77
0.72
0.6
0.82
0.79
SControl
0.84
0.85
0.7
0.17
0.87
0.89
0.89
0.64
0.64
0.23
0.17
0.67
0.93
0.67
0.61
0.63
0.19
0.28
0.71
0.52
0.75
0.69
0.71
0.01
0
0.82
0.68
0.85
For each dataset, the numbers in the first row are rand index, those in the second row are purity, those in the third row are Fmeasure, and those in the forth row are NMI.
