Table 1 Statistics of the networks studied in this paper.

From: A perturbation-based framework for link prediction via non-negative matrix factorization

Network

NN

NE

LD

AD

APL

C

CC

P

LCP-corr

C. elegans

297

2148

0.0489

14.4646

2.4553

0.0014

0.2924

−0.1632

0.9056

Email

1133

5451

0.0085

9.6222

3.6060

0.0002

0.1663

0.0782

0.8538

Karate

34

78

0.1390

4.5882

2.4082

0.0129

0.5706

−0.4756

0.7562

Word

112

425

0.0684

7.5893

2.5356

0.0036

0.1728

−0.1293

0.8528

Jazz

198

2742

0.1406

27.6970

2.2350

0.0023

0.6175

0.0202

0.9485

USAir

332

2126

0.0387

12.8072

2.7381

0.0011

0.6252

−0.2079

0.9799

Yeast

2361

6646

0.0024

5.6298

5.0960

0.0001

0.3057

0.4539

0.9686

PB

1222

16714

0.0224

27.3552

2.7375

0.0003

0.3203

−0.2213

0.9286

NS

379

914

0.0128

4.8232

6.0419

0.0004

0.7412

−0.0817

0.9224

Power

4941

6594

0.0005

2.6691

18.9892

0.0000

0.1032

0.0035

0.8456

Router

5022

6258

0.0005

2.4922

6.4488

0.0000

0.0303

−0.1384

0.8067

Baydry

128

2106

0.2591

32.9063

1.7724

0.0045

0.3346

−0.1044

0.9112

School

69

220

0.09378

6.3768

2.965

0.005

0.4606

0.0141

0.9005

SmaGri

1024

4916

0.0094

9.6016

2.9814

0.0003

0.3071

−0.1925

0.9463

SW

233

994

0.03678

8.5322

2.3711

0.0018

0.5564

−0.3027

0.9436

  1. Where, NN, NE, LD, AD, APL, CC and P are the number of nodes, the number of edges, the link density, average degree, average shortest path length, clustering coefficient, and pearson assortative coefficient of the network, respectively. C is the average closeness for all the pair nodes of the network, LCP-corr is the correlation between LCP and CN indices presented in ref. 11.
Back to article page