Dear Graph Tool Community, I am interested in establishing differences between large undirected networks with multiple edge weights. I have been doing so thus far by using functions like similarity, passing individual edge weights to review effects of a given factor on graphical similarity. But, I am especially interested in comparing generative community based structure acquired with the nested SBM generated by these different edge weights. For instance, comparing the results of a nested SBM generated with 3 edge weights, and another with 3 different edge weights, where the vertices are all the same. However, I’m not sure the best way to implement this. Does anyone have some advice how best to go about it? (The only thought that comes to mind is compute a dice score for a given comparison, but not sure this is the best way…?) Thank you for your time. James
Am 14.02.20 um 12:20 schrieb James Ruffle:
For instance, comparing the results of a nested SBM generated with 3 edge weights, and another with 3 different edge weights, where the vertices are all the same. However, I’m not sure the best way to implement this. Does anyone have some advice how best to go about it?
You can compare the partitions at the lowest level using variation of information, for example. -- Tiago de Paula Peixoto <tiago@skewed.de>
Excellent - many thanks. James
On 18 Feb 2020, at 17:58, Tiago de Paula Peixoto <tiago@skewed.de> wrote:
Am 14.02.20 um 12:20 schrieb James Ruffle:
For instance, comparing the results of a nested SBM generated with 3 edge weights, and another with 3 different edge weights, where the vertices are all the same. However, I’m not sure the best way to implement this. Does anyone have some advice how best to go about it?
You can compare the partitions at the lowest level using variation of information, for example.
-- Tiago de Paula Peixoto <tiago@skewed.de> _______________________________________________ graph-tool mailing list graph-tool@skewed.de https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.skewed.de%2Fmailman%2Flistinfo%2Fgraph-tool&data=02%7C01%7C%7C511f9db4ec6e4e56f60508d7b49c2f7c%7C569df091b01340e386eebd9cb9e25814%7C0%7C0%7C637176455225434130&sdata=7xaBY1YUxT0RyWHjnZGuOUQ7gZx6TwgelLCArVuUBvI%3D&reserved=0
participants (2)
-
James Ruffle -
Tiago de Paula Peixoto