This is the bipartite picture tagging network of Left nodes represent tags and right nodes represents pictures. An edge connects a picture with an assigned tag.


Internal namepics_ti tag–item
Data source
AvailabilityDataset is available for download
Consistency checkDataset passed all tests
Feature network
Node meaningTag, picture
Edge meaningAssignment
Network formatBipartite, undirected
Edge typeUnweighted, multiple edges


Size n =577,437
Left size n1 =82,035
Right size n2 =495,402
Volume m =2,298,816
Unique edge count m̿ =1,800,330
Wedge count s =30,601,838,083
Claw count z =1,985,799,614,059,094
Square count q =2,199,351,548
4-Tour count T4 =140,005,809,036
Maximum degree dmax =405,429
Maximum left degree d1max =405,429
Maximum right degree d2max =790
Average degree d =7.962 14
Average left degree d1 =28.022 4
Average right degree d2 =4.640 30
Fill p =4.429 91 × 10−5
Average edge multiplicity m̃ =1.276 89
Size of LCC N =564,218
Diameter δ =17
50-Percentile effective diameter δ0.5 =3.468 93
90-Percentile effective diameter δ0.9 =4.915 48
Median distance δM =4
Mean distance δm =3.887 09
Gini coefficient G =0.764 600
Balanced inequality ratio P =0.201 174
Left balanced inequality ratio P1 =0.083 396 4
Right balanced inequality ratio P2 =0.282 309
Tail power law exponent γt =2.591 00
Degree assortativity ρ =−0.167 621
Degree assortativity p-value pρ =0.000 00
Spectral norm α =1,548.34
Algebraic connectivity a =0.009 718 28
Spectral separation 1[A] / λ2[A]| =3.644 49
Controllability C =442,519
Relative controllability Cr =0.766 350


Fruchterman–Reingold graph drawing

Degree distribution

Cumulative degree distribution

Lorenz curve

Spectral distribution of the adjacency matrix

Spectral distribution of the normalized adjacency matrix

Spectral distribution of the Laplacian

Spectral graph drawing based on the adjacency matrix

Spectral graph drawing based on the Laplacian

Spectral graph drawing based on the normalized adjacency matrix

Degree assortativity

Zipf plot

Hop distribution

Edge weight/multiplicity distribution

Matrix decompositions plots



[1] Jérôme Kunegis. KONECT – The Koblenz Network Collection. In Proc. Int. Conf. on World Wide Web Companion, pages 1343–1350, 2013. [ http ]
[2] Nicolas Neubauer and Klaus Obermayer. Analysis of the folksonomy. Unpublished, 2010.