Community detection

Table Of Contents

Next topic

community API

This Page

Community detection for NetworkX’s documentation

This module implements community detection.

It uses the louvain method described in Fast unfolding of communities in large networks, Vincent D Blondel, Jean-Loup Guillaume, Renaud Lambiotte, Renaud Lefebvre, Journal of Statistical Mechanics: Theory and Experiment 2008(10), P10008 (12pp)

It depends on Networkx to handle graph operations : http://networkx.lanl.gov/

The program itself can be found at

http://perso.crans.org/aynaud/communities/community.py

Example :

As a classical software :

You should consider using the cpp version at http://findcommunities.googlepages.com/ !

./community.py file.bin > tree

where file.bin is a binary graph as generated by the convert utility of the cpp version.

You can after that use the generated file with the hierarchy utility of the cpp version. Note that the program does not make much verifications about the arguments, and is expecting a friendly use.

As python module :

import community
import networkx as nx
import matplotlib.pyplot as plt

#better with karate_graph() as defined in networkx example.
#erdos renyi don't have true community structure
G = nx.erdos_renyi_graph(30, 0.05)

#first compute the best partition
partition = community.best_partition(G)

#drawing
size = float(len(set(partition.values())))
pos = nx.spring_layout(G)
count = 0.
for com in set(partition.values()) :
    count = count + 1.
    list_nodes = [nodes for nodes in partition.keys()
                                if partition[nodes] == com]
    nx.draw_networkx_nodes(G, pos, list_nodes, node_size = 20,
                                node_color = str(count / size))


nx.draw_networkx_edges(G,pos, alpha=0.5)
plt.show()

Changelog :

  • 01/14/2010 : modification to use networkx 1.01 graph api and adding the possibility to start the algorithm with a given partition
  • 04/10/2009 : increase of the speed of the detection by caching node degrees

License :

Copyright (C) 2009 by Thomas Aynaud <thomas.aynaud@lip6.fr>

Distributed under the terms of the GNU Lesser General Public License http://www.gnu.org/copyleft/lesser.html

Indices and tables