Added Stochastic Variability in Community Detection Algorithms#820
Added Stochastic Variability in Community Detection Algorithms#820SKG24 wants to merge 7 commits intoigraph:mainfrom
Conversation
This example demonstrates the variability of stochastic community detection methods by analyzing the consistency of multiple partitions using similarity measures (NMI, VI, RI) on both random and structured graphs.
|
I won't have time to look in detail today, but I checked whether the docs build with this change, and unfortunately they do not. Can you please check if you can fix this? You can build the docs using |
|
Issue Description: AttributeError: module 'igraph' has no attribute '_igraph'. Did you mean: 'Graph'?
|
|
I think I've fixed the build issue in the |
|
Thank you! It worked. Screen.Recording.2025-03-27.at.9.33.28.PM.mov |
szhorvat
left a comment
There was a problem hiding this comment.
This is a nice illustration!
I left a few comments for improvement.
Please remove changes to the sg_execution_times.rst file. This file was likely committed by accident, and I think we should remove it (but not as part of this PR).
I have made the changes as per the review.
|
I have made the suggested changes. Screen.Recording.2025-03-28.at.3.38.52.PM.mov |
Should I directly delete this file from PR through files changed section? |
Yes, that would be fine. |
|
This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 14 days if no further activity occurs. Thank you for your contributions. |
|
Continued in #840. I can't push to this as the work was done on the |

Used Chatgpt for understanding the functions and theory around the mathematical process.
At first run:

At second run:

This supports stochastic method might give wildly different answers on a network without any significant community structure, while it gives consistent answers on one that has obvious communities.
Since the number of iterations is set to 50, even in structured graphs, slight variations in similarity scores may be observed.