Wikipedia talk (en)
This is the communication network of the English 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 = | 2,987,535
|
Volume | m = | 24,981,163
|
Unique edge count | m̿ = | 9,379,561
|
Loop count | l = | 5,655,527
|
Wedge count | s = | 57,066,712,805
|
Claw count | z = | 1,510,569,161,023,742
|
Cross count | x = | 4.396 49 × 1019
|
Triangle count | t = | 41,915,754
|
Square count | q = | 22,498,726,804
|
Maximum degree | dmax = | 488,182
|
Maximum outdegree | d+max = | 488,169
|
Maximum indegree | d−max = | 121,250
|
Average degree | d = | 16.723 6
|
Fill | p = | 1.050 89 × 10−6
|
Average edge multiplicity | m̃ = | 2.663 36
|
Size of LCC | N = | 2,859,574
|
Size of LSCC | Ns = | 249,610
|
Relative size of LSCC | Nrs = | 0.083 550 5
|
Diameter | δ = | 9
|
50-Percentile effective diameter | δ0.5 = | 3.233 96
|
90-Percentile effective diameter | δ0.9 = | 3.877 48
|
Median distance | δM = | 4
|
Mean distance | δm = | 3.658 54
|
Gini coefficient | G = | 0.899 562
|
Balanced inequality ratio | P = | 0.109 987
|
Outdegree balanced inequality ratio | P+ = | 0.073 885 3
|
Indegree balanced inequality ratio | P− = | 0.160 171
|
Relative edge distribution entropy | Her = | 0.785 090
|
Power law exponent | γ = | 2.827 07
|
Tail power law exponent | γt = | 1.811 00
|
Degree assortativity | ρ = | −0.096 231 0
|
Degree assortativity p-value | pρ = | 0.000 00
|
Clustering coefficient | c = | 0.002 203 51
|
Directed clustering coefficient | c± = | 0.017 955 2
|
Spectral norm | α = | 90,816.2
|
Operator 2-norm | ν = | 45,410.1
|
Cyclic eigenvalue | π = | 45,406.0
|
Spectral separation | |λ1[A] / λ2[A]| = | 1.271 98
|
Reciprocity | y = | 0.214 524
|
Non-bipartivity | bA = | 0.829 994
|
Normalized non-bipartivity | bN = | 0.031 296 9
|
Algebraic non-bipartivity | χ = | 0.064 187 3
|
Spectral bipartite frustration | bK = | 0.002 707 63
|
Controllability | C = | 2,488,813
|
Relative controllability | Cr = | 0.833 066
|
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.
|