Two algorithms for finding modularity in bipartite networks are presented: LPAwb+ and DIRTLPAwb+.
These are based on the LPAb+ algorithm of Liu & Murata, 2010 for use on bipartite/two-mode networks. The algorithm has been modified such that the weighted modularity of a network can be found (note that the result is equivalent to that found by the LPAb+ algorithm if the network is binary). Code is currently available for Julia, MATLAB/Octave and R langauges.
For details of the methods please view the draft paper (in the 'paper' directory), or in Beckett, 2016.
Language specific instructions are included with the code. Source the relevant code in your favourite language; then run
LPA_wb_plus(MATRIX) # find labels and weighted modularity using LPAwb+
where MATRIX is the incidence/biadjacency matrix describing the input network. Three outputs are returned: redlabels - the module labels for each row in the input matrix, bluelabels - the module labels for each row in the input matrix and Q - the modularity score.
Code for plotting modular structure is also provided. Please view the REAME files within the relevant code folder to learn more about computing modularity using LPAwb+ and DIRTLPAwb+.
The paper describing these algoirithms is now published:
Beckett S.J. 2016. Improved community detection in weighted bipartite networks. Royal Society Open Science 3: 140536.
For details of the LPAb+ algorithm please see Liu & Murata, 2010 .
Liu X., Murata T. 2010. An Efficient Algorithm for Optimizing Bipartite Modularity in Bipartite Networks. Journal of Advanced Computational Intelligence and Intelligent Informatics (JACIII) 14(4): 408-415.
This repository has been archived using Zenodo and has been assigned DOI: 10.5281/zenodo.34055. If you use LPAwb+ or DIRTLPAwb+ please cite this repository or the accompanying paper when it becomes available.
Click here to view notes of changes between release versions.
This package is authored by Stephen Beckett (@BeckettStephen). If you have any questions please contact me!