best way to identify groups in weighted graph
Good evening, My interrogation is quite vague… Also the answer is clearly data dependant and so it may be irritating. If so I apologize for it. I am trying to find a suitable way to identify groups in a graph with nodes connected with weighted edges. As I want the process to be as “flexible“ as possible I planed to use MCL algorithm (https://micans.org/mcl/). Also I planed to neglect weak weghted edges (with an offset) befor perfoming the calculation an set a limite to the number of Expantion/Inflation process... I wonder if you have any advice better idea or warning with this ? Thanks / Regards -- Sent from: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/
On 24.04.2018 16:32, M. Vigouroux wrote:
I wonder if you have any advice better idea or warning with this ?
The MCL algorithm is not implemented in graph-tool. What is implemented is statistical inference approach based on weighted stochastic block models as explained in the documentation: https://graph-tool.skewed.de/static/doc/demos/inference/inference.html#edge-... -- Tiago de Paula Peixoto <tiago@skewed.de>
Thank you, I will read this closely Have a nice day -- Sent from: http://main-discussion-list-for-the-graph-tool-project.982480.n3.nabble.com/
participants (2)
-
M. Vigouroux -
Tiago de Paula Peixoto