You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For the moment I have no plan to adapt to directed graphs.
Modularity meaning is not very clear for a directed graph and I do not know a "standard" implementation of Louvain algorithms on such graph. The usual way is to consider directed graph as undirected.
Activity
taynaud commentedon Oct 26, 2018
Hello,
For the moment I have no plan to adapt to directed graphs.
Modularity meaning is not very clear for a directed graph and I do not know a "standard" implementation of Louvain algorithms on such graph. The usual way is to consider directed graph as undirected.
Best
deemeetree commentedon Mar 31, 2019
Hello @taynaud — could you please elaborate on why modularity meaning is not very clear for directed graphs? What is the controversy here?
Do you know of any methods for reliable community detection and modularity calculation on directed graphs?
wilmeragsgh commentedon Oct 10, 2019
There's a way presented here: https://github.com/nicolasdugue/DirectedLouvain but they do not provide python implementation/interface to its solution.
They also made a few changes on the modularity function the algorithm use, and they have a paper about it: https://hal.archives-ouvertes.fr/hal-01231784
DonaldTsang commentedon Dec 10, 2019
would like to see more on this
taynaud commentedon Apr 6, 2020
Related issue: #58