Amazon (MDS)
This is the copurchase network of Amazon based on the "customers who bought
this also bought" feature. Nodes are products and an undirected edge between
two nodes shows that the corresponding products have been frequently bought
together.
Metadata
Statistics
Size  n =  334,863

Volume  m =  925,872

Wedge count  s =  9,752,186

Claw count  z =  142,823,893

Cross count  x =  6,722,504,872

Triangle count  t =  667,129

Square count  q =  3,125,323

4Tour count  T_{4} =  65,863,072

Maximum degree  d_{max} =  549

Average degree  d =  5.529 86

Fill  p =  1.651 38 × 10^{−5}

Size of LCC  N =  334,863

Diameter  δ =  47

50Percentile effective diameter  δ_{0.5} =  11.076 0

90Percentile effective diameter  δ_{0.9} =  14.851 4

Mean distance  δ_{m} =  11.725 3

Gini coefficient  G =  0.385 905

Relative edge distribution entropy  H_{er} =  0.976 681

Power law exponent  γ =  1.691 22

Tail power law exponent  γ_{t} =  3.581 00

Tail power law exponent with p  γ_{3} =  3.581 00

pvalue  p =  0.022 000 0

Degree assortativity  ρ =  −0.058 819 6

Degree assortativity pvalue  p_{ρ} =  0.000 00

Clustering coefficient  c =  0.205 224

Spectral norm  α =  23.975 6

Nonbipartivity  b_{A} =  0.031 835 0

Normalized nonbipartivity  b_{N} =  0.012 895 0

Spectral bipartite frustration  b_{K} =  0.001 185 21

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]

Jaewon Yang and Jure Leskovec.
Defining and evaluating network communities based on groundtruth.
In Proc. ACM SIGKDD Workshop on Min. Data Semant., page 3,
2012.
