[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [igraph] edge_betweenness community detection

From: Tamas Nepusz
Subject: Re: [igraph] edge_betweenness community detection
Date: Fri, 15 Jan 2010 14:56:33 +0000
User-agent: Mutt/1.5.20 (2009-06-14)

> I run igraph_community_edge_betweenness () on my dataset. The dataset
> consists of 12,000 nodes and 222,156 edges.
> The program has been running the whole night. It still does not finish.
> In contrast, fastgreedy community detection can finish in much less time.
> So, could anyone can tell me a rough time that the edge_betweenness() needs
> to detect community?
A rough estimate is something like "a LOT". :)

The edge betweenness method calculates the betweennesses of ALL the
edges in EVERY step of the algorithm. The algorithm will take 222.156
steps since it removes edges one by one. So, a quick estimate can be
given by doing the following:

1. Measure the time it takes to calculate all the edge betweennesses
(222.156 edges) in the network.

2. Multiply it by, say, half the number of edges. This gives you a rough


reply via email to

[Prev in Thread] Current Thread [Next in Thread]