Re: [igraph] Slow community detection

[Tamas Nepusz]

Hi,

> In the first case the Rgui freezed out, but in the second case the community
> detection was done in less than 5 minutes. So, it looks like the comm. det.
> algorithms can't handle the big directed graphs.
Label propagation is not guaranteed to converge for directed graphs
(although sometimes it does), and it was never meant for directed
graphs. Imagine a directed ring consisting of N nodes such that node i
points to node (i+1) modulo N. In that case, labels would simply
propagate forward along the edges ad infinitum.

> Do you have any idea what is wrong with the Python igraph package or what I
> should do to recreate the experiment in Python?
You have installed the wrong package; the Python interface of igraph
is called "python-igraph" and not simply "igraph"; that one is a
different package. This is the correct one:

https://pypi.python.org/pypi/python-igraph/

T.