Edinburgh Associative Thesaurus

This is the Edinburgh Associative Thesaurus. Nodes are English words, and a directed link from A to B denotes that the word B was given as a response to the stimulus word A in user experiments. Multiple links are allowed, and the multiplicity of an edge denotes the number of times a word has been given as a response to another word. In some case, the given word itself was given as a response, leading to loops in the network.


Internal nameeat
NameEdinburgh Associative Thesaurus
Data sourcehttp://vlado.fmf.uni-lj.si/pub/networks/data/dic/eat/Eat.htm
AvailabilityDataset is available for download
Consistency checkDataset passed all tests
Lexical network
Node meaningWord
Edge meaningAssociation
Network formatUnipartite, directed
Edge typeUnweighted, multiple edges
ReciprocalContains reciprocal edges
Directed cyclesContains directed cycles
LoopsContains loops


Size n =23,132
Volume m =511,764
Unique edge count m̿ =312,310
Wedge count s =30,388,904
Claw count z =2,967,734,991
Cross count x =376,129,003,421
Triangle count t =409,174
Square count q =16,682,968
4-Tour count T4 =255,613,548
Maximum degree dmax =1,076
Maximum outdegree d+max =78
Maximum indegree dmax =1,044
Average degree d =44.247 3
Fill p =0.000 583 659
Average edge multiplicity m̃ =1.638 64
Size of LCC N =23,132
Size of LSCC Ns =7,751
Relative size of LSCC Nrs =0.335 077
Diameter δ =6
50-Percentile effective diameter δ0.5 =2.935 81
90-Percentile effective diameter δ0.9 =3.852 38
Median distance δM =3
Mean distance δm =3.431 44
Gini coefficient G =0.696 577
Relative edge distribution entropy Her =0.911 566
Power law exponent γ =1.469 67
Tail power law exponent γt =3.211 00
Degree assortativity ρ =−0.047 673 9
Degree assortativity p-value pρ =0.000 00
In/outdegree correlation ρ± =+0.532 290
Clustering coefficient c =0.040 393 8
Spectral norm α =189.271
Operator 2-norm ν =131.648
Cyclic eigenvalue π =57.307 7
Algebraic connectivity a =0.961 452
Spectral separation 1[A] / λ2[A]| =1.526 69
Reciprocity y =0.095 674 2
Non-bipartivity bA =0.542 479
Normalized non-bipartivity bN =0.482 007
Spectral bipartite frustration bK =0.007 915 30
Controllability C =14,922
Relative controllability Cr =0.645 080


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

In/outdegree scatter plot

Edge weight/multiplicity distribution

Clustering coefficient distribution

Average neighbor degree 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] George R. Kiss, Christine Armstrong, and Robert Milroy. An associative thesaurus of English and its computer analysis. The Comput. and Literary Studies, 1973.