marblet / BiMLPA

Implementation of "BiMLPA: Community Detection in Bipartite Networks by Multi-Label Propagation"

BiMLPA

"BiMLPA: Community Detection in Bipartite Networks by Multi-Label Propagation", NetSci-X 2020 https://link.springer.com/chapter/10.1007/978-3-030-38965-9_2

This is the implementation of BiMLPA. BiMLPA is to detect the many-to-many correspondence community in bipartite networks using multi-label propagation algorithm. 日本語版READMEはこちら

Installation

In order to install the package just download (or clone) the current project and copy the demon folder in the root of your application.

Alternatively use pip:

sudo pip install bimlpa

If you like to install the latest version of the package from the repository use:

sudo pip install git+https://github.com/marblet/BiMLPA

Usage

from BiMLPA import *
import networkx as nx

G = generate_network('BiMLPA/test/southernwomen.net')

# The parameters are set to theta=0.3, lambda=7
bimlpa = BiMLPA_SqrtDeg(G, 0.3, 7)
bimlpa.start()
relabeling(G)
top_coms, bottom_coms = output_community(G)

# If the community structure is known, the normalized mutual information score can be calculated
# using calc_NMI by assigning the correct community number to the attribute 'community' of the node.
community = {i+1: 0 for i in range(9)}
community.update({i+10: 1 for i in range(9)})
community.update({i+19: 2 for i in range(6)})
community.update({i+25: 3 for i in range(3)})
community.update({i+28: 4 for i in range(5)})
nx.set_node_attributes(G, name='community', values=community)

print('NMI : ', calc_NMI(G))