How do I run the louvain community detection algorithm in igraph?

Question

Could someone please provide me with a simple example of how to run the louvain community detection algorithm in igraph using the python interface. Is there any documentation?

Thanks!

Answer 1

It's called multilevel.community .

According to https://bugs.launchpad.net/igraph/+bug/925038 ... this functionality does exist it's just called igraph_community_multilevel

If you look in the github repository for igraph

https://github.com/igraph/igraph/blob/master/src/community.c

igraph_community_multilevel does exist and it's written in C. I'm not 100% positive this is the algorithm you want but it might be.

This is great news! Thanks! Is this functionality exported into R? Why does the function bear a generic name (igraph_community_multilevel) instead of the name which the authors gave is ("louvain method")? Using the name "louvain" would make it easier for the users to find the function!

Answer 2

Here is how to estimate the modularity Q using louvain algorithm in 3 different modules in python (igraph,networkx,bct).

import numpy as np
import networkx as nx
np.random.seed(9)

# I will generate a stochastic block model using `networkx` and then extract the weighted adjacency matrix.
sizes = [50, 50, 50] # 3 communities
probs = [[0.25, 0.05, 0.02],
         [0.05, 0.35, 0.07],
         [0.02, 0.07, 0.40]]

# Build the model
Gblock = nx.stochastic_block_model(sizes, probs, seed=0)

# Extract the weighted adjacency
W = np.array(nx.to_numpy_matrix(Gblock, weight='weight'))
W[W==1] = 1.5
print(W.shape)
# (150, 150)

#* Modularity estimation using Louvain algorithm 
# 1. `igraph` package

from igraph import *
graph = Graph.Weighted_Adjacency(W.tolist(), mode=ADJ_UNDIRECTED, attr="weight", loops=False)
louvain_partition = graph.community_multilevel(weights=graph.es['weight'], return_levels=False)
modularity1 = graph.modularity(louvain_partition, weights=graph.es['weight'])
print("The modularity Q based on igraph is {}".format(modularity1))

# 2. `networkx`package using `python-louvain`
# https://python-louvain.readthedocs.io/en/latest/

import networkx as nx
import community
G = nx.from_numpy_array(W)
louvain_partition = community.best_partition(G, weight='weight')
modularity2 = community.modularity(louvain_partition, G, weight='weight')
print("The modularity Q based on networkx is {}".format(modularity2))

# 3. `bct` module
# https://github.com/aestrivex/bctpy

import bct
com, q = bct.community_louvain(W)
print("The modularity Q based on bct is {}".format(q))

This prints:

The modularity Q based on igraph is 0.4257613861340037
The modularity Q based on networkx is 0.4257613861340036
The modularity Q based on bct is 0.42576138613400366