Implement louvain method for community detection

[sds]

The louvaine community detection algorithm
https://sites.google.com/site/findcommunities/
scales better than the fast/greedy and avoids supercommunities

[Tamás Nepusz (ntamas)]

The development version already includes the Louvain method (see igraph_community_multilevel in the 0.6 branch). This is not released yet but you can compile it yourself. Nightly tarballs are available at http://code.google.com/p/igraph

[sds (sds-gnu)]

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!

[Tamás Nepusz (ntamas)]

Yes, you can install the R development version from the nightlies and it should be there; I think it's called "multilevel.community".

Re the naming: well, the sword has two edges -- if we named it like "igraph_community_louvain", users looking for a generic multi-level community algorithm would not be able to find it ;) Since all the other community detection functions use generic names (e.g., "igraph_community_edge_betweenness" instead of "igraph_community_girvan_newman", "igraph_community_fastgreedy" instead of "igraph_community_clauset_newman_moore", "igraph_community_leading_eigenvector" instead of "igraph_community_newman" etc), we decided to stick to this naming scheme.

[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/448