Bipartite modularity

[Gábor Csárdi]

This means using another null model, in the general modularity formula. See
Michael J Barber: Modularity and community detection in bipartite networks,
Phys Rev E 76 066102
http://pre.aps.org/abstract/PRE/v76/i6/e066102

[Tamás Nepusz (ntamas)]

Note that there are several alternative definitions for bipartite modularity, so we may have to be prepared for those as well. E.g.:

R. Guimerà, M. S.- Pardo, and L. A. N. Amaral, “Module identification in bipartite and directed networks,” Phys. Rev. E 76, 036102, 2007.
http://arxiv.org/abs/physics/0701151

And there's also this one:
http://portal.acm.org/citation.cfm?id=1557914.1557956

[Tamás Nepusz (ntamas)]

By the way, there are several other modularity variants that all differ in the null model being used; for instance, a directed modularity measure was proposed by Kim et al based on using the PageRank vectors instead of the degrees. There's also a Laplacian formulation of modularity by Lambiotte, Delvenne and Barahona: http://arxiv.org/abs/0812.1770. We might want to adapt the modularity function completely so it can take different null models.

[Gábor Csárdi (gabor.csardi)]

Yes, I know, but I like the Barber paper better, so I'll implement that at some point. Feel free to implement the rest or redesign the modularity API if you feel like it.

[Gábor Csárdi (gabor.csardi)]

The development of igraph has moved to github, so please do not comment on this bug here. You are of course welcome to comment on github, here:
https://github.com/igraph/igraph/issues/194