Enron (clean)
The Enron email network consists of 1,148,072 emails sent between employees of
Enron between 1999 and 2003. Nodes in the network are individual employees and
edges are individual emails. It is possible to send an email to oneself, and
thus this network contains loops. In this version of the network, nodes with
erroneous timestamps were removed.
Metadata
Statistics
Size  n =  87,273

Volume  m =  1,147,126

Unique edge count  m̿ =  321,288

Loop count  l =  13,080

Wedge count  s =  49,130,407

Claw count  z =  13,799,850,890

Cross count  x =  3,784,732,758,298

Triangle count  t =  1,178,289

Square count  q =  92,655,137

4Tour count  T_{4} =  938,356,410

Maximum degree  d_{max} =  38,778

Maximum outdegree  d^{+}_{max} =  32,613

Maximum indegree  d^{−}_{max} =  6,165

Average degree  d =  26.288 2

Fill  p =  4.234 95 × 10^{−5}

Average edge multiplicity  m̃ =  3.570 40

Size of LCC  N =  84,220

Size of LSCC  N_{s} =  9,160

Relative size of LSCC  N^{r}_{s} =  0.104 958

Diameter  δ =  14

50Percentile effective diameter  δ_{0.5} =  4.383 96

90Percentile effective diameter  δ_{0.9} =  5.794 71

Median distance  δ_{M} =  5

Mean distance  δ_{m} =  4.890 80

Gini coefficient  G =  0.907 479

Balanced inequality ratio  P =  0.100 665

Outdegree balanced inequality ratio  P_{+} =  0.106 781

Indegree balanced inequality ratio  P_{−} =  0.134 847

Relative edge distribution entropy  H_{er} =  0.827 659

Power law exponent  γ =  2.650 47

Tail power law exponent  γ_{t} =  1.761 00

Degree assortativity  ρ =  −0.166 816

Degree assortativity pvalue  p_{ρ} =  0.000 00

In/outdegree correlation  ρ^{±} =  +0.403 460

Clustering coefficient  c =  0.071 948 7

Directed clustering coefficient  c^{±} =  0.111 155

Spectral norm  α =  7,808.15

Operator 2norm  ν =  3,904.14

Cyclic eigenvalue  π =  3,904.00

Algebraic connectivity  a =  0.004 229 27

Reciprocity  y =  0.146 678

Nonbipartivity  b_{A} =  0.622 351

Normalized nonbipartivity  b_{N} =  0.001 262 22

Spectral bipartite frustration  b_{K} =  8.929 14 × 10^{−5}

Plots
Matrix decompositions plots
Downloads
References
[1]

Jérôme Kunegis.
KONECT – The Koblenz Network Collection.
In Proc. Int. Conf. on World Wide Web Companion, pages
1343–1350, 2013.
[ http ]

[2]

Bryan Klimt and Yiming Yang.
The Enron corpus: A new dataset for email classification research.
In Proc. Eur. Conf. on Mach. Learn., pages 217–226, 2004.
