Show HN: KarateClub a Python library for unsupervised machine learning on graphs(github.com)

102 pointsby benitorosenbergvor 6 Jahren9 comments

Show HN: KarateClub a Python library for unsupervised machine learning on graphs

https://github.com/benedekrozemberczki/karatecluB

3 comments

cl42vor 6 Jahren

This is a great name!! For those who aren't familiar, there's a famous graph data set about a karate club: https://en.wikipedia.org/wiki/Zachary%27s_karate_club

Thanks for bringing me back to undergrad... :) (...where I learned about this data set.)

lasershowvor 6 Jahren

+1 on the great name and thanks for the reference link

teruakohatuvor 6 Jahren

Thank you for sharing. The DOI is 10.2307/3629752 if anyone wants to look it up.

lammalamma25vor 6 Jahren

Very cool. This kind of work makes python and ML better for everyone. I'm not sure I have a use case at the moment, but I'll definitely be keeping an eye on this one.

fudged71vor 6 Jahren

I've recently been looking into community detection for NetworX graphs for my collection of notes. How does performance compare with other community detection packages for NetworkX like Python-Louvain?

kugestuvor 6 Jahren

Random side note: nowadays 'Leiden' is a recommended improvement on 'Louvain', fixing some connectivity issues and being faster: https://www.nature.com/articles/s41598-019-41695-z https://github.com/vtraag/leidenalg

benitorosenbergvor 6 Jahren

Performance measured by what? NMI? Modularity?

fudged71vor 6 Jahren

Not super familiar with the measures, I'm just curious how it compares to other available approaches

benitorosenbergvor 6 Jahren

Louvain greedily optimizes modularity - it is hard to beat it in terms of modularity.