Cluster transitivity

asked 2021-01-21 16:26:34 +0100

SYLA gravatar image

updated 2021-01-21 16:29:57 +0100

The cluster transitivity (the fraction of closed triangles ) of g is equal to:

C =(number of triangles) × 3/(number of connected triples)

Formula (7.28) Networks Mark Newman (2018)

I would like to ask you about the method cluster_transitivity() It gives 0.0853107962707866 for this network: , but a table from this page shows that Fraction of closed triangles is 0.03015. So, these two number are not equal.

If I omit 3 in the formula (7.28) and I get 0.0284369320902622.


On this page: (at the end), we have a definition of the fraction of closed triangles: Number of connected triples of nodes / number of (undirected) length 2 paths

edit retag flag offensive close merge delete


The method cluster_transitivity is simply

import networkx
return networkx.transitivity(self.networkx_graph())

so may be you should ask the developers of networkx the difference with the method implemented and used by SNAP ?`

David Coudert gravatar imageDavid Coudert ( 2021-01-21 22:38:38 +0100 )edit