Skip to main content

Table 2 Basic statistics of generated ERGM networks, and the population of node pairs

From: The risk of node re-identification in labeled social graphs

Network ERGM d C r κ |S| (millions)
polblogs dc 0.02 0.03 .08 2.52 5.5
  cc 0.02 0.33 -0.02 2.69 13.1
  apl 0.02 0.10 -0.06 2.49 11.5
fb-caltech dc 0.06 0.08 0.11 2.13 1.2
  cc 0.06 0.42 -0.06 2.73 4.1
  apl 0.06 0.07 0.11 1.97 1.2
fb-dartmouth dc 0.01 0.17 0.07 2.66 14.5
  cc 0.01 0.24 0.04 2.77 13.2
  apl 0.01 0.20 0.04 2.70 14.2
fb-michigan dc 0.003 0.02 0.12 3.28 38.4
  cc 0.002 0.20 0.12 3.52 39.9
  apl 0.002 0.20 0.12 3.64 38.2
pokec-1 dc 2.02E-5 0.06 -0.04 5.60 29.5
  cc 2.05E-5 0.07 -0.04 5.84 29.3
  apl 2.04E-5 0.06 -0.04 5.63 27.3
amazon-products dc 1.82E-5 0.37 -0.06 11.86 43.7
  cc 1.82E-5 0.40 -0.06 13.52 72.5
  apl 1.82E-5 0.39 -0.06 13.47 74.3
  1. Note that dc,cc and apl define the set of parameters that used to generate ERGM graphs based on assortativity (degree correlation), clustering coefficient, and average path length, respectively. We generated a total of ≈ 500 million identical and non-identical node pairs over three ERGM graph spaces of the six real social network datasets. S is the population of generated node pairs concerning a given graph topology
\