In graph mining the concept of dividing the nodes of a network into groups or communities as part of a clustering exercise is also called modularity. How well graph generators preserve modularity?
Identifiying communities may offer insight on how the network is organized.
It is not unusual for communities to overlap at their boundaries.
Finding cliques is an NP-complete problem (Bomze et al., 1999); a popular technique is the Bron–Kerbosch method (Bron and Kerbosch, 1973).
Q = 1/m (1⁄2 Sum_ij (A_ij - Pij)\delta gigj
Working with Networks by kateto
http://stackoverflow.com/questions/24255143/graph-theory-in-networkx http://stackoverflow.com/questions/29897243/graph-modularity-in-python-networkx
http://www.sixhat.net/finding-communities-in-networks-with-r-and-igraph.html
http://igraph.org/r/doc/modularity.igraph.html