Wikipedia talk (lv)
This is the communication network of the Latvian Wikipedia. Nodes represent
users, and an edge from user A to user B denotes that user A wrote a message on
the talk page of user B at a certain timestamp.
Metadata
Statistics
Size  n =  41,424

Volume  m =  73,900

Unique edge count  m̿ =  51,098

Loop count  l =  8,740

Wedge count  s =  649,358,980

Claw count  z =  7,621,516,311,302

Cross count  x =  68,032,007,206,512,256

Triangle count  t =  10,408

Square count  q =  5,154,202

4Tour count  T_{4} =  2,638,768,688

Maximum degree  d_{max} =  35,755

Maximum outdegree  d^{+}_{max} =  35,733

Maximum indegree  d^{−}_{max} =  2,220

Average degree  d =  3.567 98

Fill  p =  2.977 83 × 10^{−5}

Average edge multiplicity  m̃ =  1.446 24

Size of LCC  N =  41,278

Size of LSCC  N_{s} =  510

Relative size of LSCC  N^{r}_{s} =  0.012 311 7

Diameter  δ =  6

50Percentile effective diameter  δ_{0.5} =  1.662 40

90Percentile effective diameter  δ_{0.9} =  3.095 04

Median distance  δ_{M} =  2

Mean distance  δ_{m} =  2.355 89

Gini coefficient  G =  0.709 842

Balanced inequality ratio  P =  0.218 606

Outdegree balanced inequality ratio  P_{+} =  0.057 659 0

Indegree balanced inequality ratio  P_{−} =  0.357 673

Relative edge distribution entropy  H_{er} =  0.630 515

Power law exponent  γ =  9.150 63

Tail power law exponent  γ_{t} =  3.291 00

Tail power law exponent with p  γ_{3} =  3.291 00

pvalue  p =  0.000 00

Outdegree tail power law exponent with p  γ_{3,o} =  2.181 00

Outdegree pvalue  p_{o} =  0.000 00

Indegree tail power law exponent with p  γ_{3,i} =  3.381 00

Indegree pvalue  p_{i} =  0.000 00

Degree assortativity  ρ =  −0.593 242

Degree assortativity pvalue  p_{ρ} =  0.000 00

In/outdegree correlation  ρ^{±} =  +0.604 761

Clustering coefficient  c =  4.808 43 × 10^{−5}

Directed clustering coefficient  c^{±} =  0.019 684 1

Spectral norm  α =  1,931.81

Operator 2norm  ν =  981.703

Cyclic eigenvalue  π =  943.056

Algebraic connectivity  a =  0.080 124 8

Spectral separation  λ_{1}[A] / λ_{2}[A] =  1.419 92

Reciprocity  y =  0.040 451 7

Nonbipartivity  b_{A} =  0.902 004

Normalized nonbipartivity  b_{N} =  0.011 472 6

Algebraic nonbipartivity  χ =  0.026 898 5

Spectral bipartite frustration  b_{K} =  0.002 753 43

Controllability  C =  40,325

Relative controllability  C_{r} =  0.973 469

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]

Jun Sun, Jérôme Kunegis, and Steffen Staab.
Predicting user roles in social networks using transfer learning with
feature transformation.
In Proc. ICDM Workshop on Data Min. in Netw., 2016.
