Skip to main content

Table 1 Some statistics for ground-truth communities from SNAP

From: Hierarchical community detection via rank-2 symmetric nonnegative matrix factorization

Dataset

#Nodes

#Edges

Nodes that belong to

Nodes that belong to

0 Community

1 Community

Count

%

Count

%

Rel %

DBLP06

317,080

1,049,866

56,082

17.69

150,192

47.37

57.55

Youtube

1,134,890

2,987,624

1,082,215

95.36

32,613

2.87

61.91

Amazon

334,863

925,872

14,915

4.45

10,604

3.17

3.31

LiveJournal

3,997,962

34,681,189

2,850,014

71.29

394,234

9.86

34.34

Friendster

65,608,366

1,806,067,135

57,663,417

87.89

3,546,017

5.40

44.63

Orkut

3,072,441

117,185,083

750,142

24.42

128,094

4.17

5.52

  1. The last few columns show the number of nodes that do not belong to any communities and the number of nodes that belong to only one community. The “Rel %” is the number of nodes that belong to one community divided by the number of nodes that belong to at least one community