Skip to main content
EPJ Data Science

Table 3 Comparison of the clustering results for the users with degree \(k >100\) based on the k-mean and \(H/T\) break algorithm for the SVEIN and SVCN. N and f represents the total number and the percentage of users. \(n_{k}\) stands for the cumulative number of users in the k-th layer. \(\langle r \rangle \) is the average scale ratio

From: Comparative analysis of layered structures in empirical investor networks and cellphone communication networks

 

N

f

\(n_{1}\)

\(n_{2}\)

\(n_{3}\)

\(n_{4}\)

\(n_{5}\)

\(n_{6}\)

〈r〉

Panel A: Clustering results of SVEIN

 

k-means

         

c = 2

114

27.9%

27.8

121.7

    

3.84

c = 3

176

43.0%

10.9

45.8

141.7

   

3.04

c = 4

119

29.1%

5.4

20.8

57.5

151.4

  

2.64

H/T break

         

c = 4

54

11.3%

2.9

11.0

37.1

133.1

  

3.45

c = 5

273

56.9%

1.6

5.3

15.0

42.8

133.0

 

3.00

c = 6

153

31.9%

1.2

3.2

7.5

18.5

51.3

156.0

2.88

Panel B: Clustering results of SVCN

 

k-means

         

c = 4

16,918

41.1%

3.0

12.8

42.8

132.0

  

3.22

c = 5

15,209

36.9%

2.1

7.3

20.4

54.2

141.4

 

2.66

c = 6

9049

22.0%

1.6

5.1

12.5

28.9

66.5

154.0

2.33

H/T break

         

c = 3

3308

5.7%

5.0

27.1

126.7

   

4.71

c = 4

29,125

50.2%

2.1

8.7

33.4

133.9

  

3.97

c = 5

25,539

44.1%

1.2

3.8

11.7

39.5

147.6

 

3.61

Zhou

  

5

15

50

150

  

3.00

Back to article page