This repo is the authors' implementation of the GNN-style algorithm for community detection described in the paper "Graph Neural Network Inspired Algorithm for Unsupervised Network Community Detection".
This paper proposes a new variant of the recurrent graph neural network algorithm for unsupervised network community detection through modularity optimization. The new algorithm's performance is compared against state-of-the-art methods. The approach also serves as a proof-of-concept for the broader application of recurrent graph neural networks to unsupervised network optimization.
GNNS.ipynb notebook is ready to be run in Google Colab. Just follow the link: https://colab.research.google.com/github/Alexander-Belyi/GNNS/blob/master/GNNS.ipynb. It should reproduce the results presented in the paper (enable the GPU backend to reproduce the running time figures).
If you find this work useful, please, consider citing:
S. Sobolevsky, A. Belyi, "Graph Neural Network Inspired Algorithm for Unsupervised Network Community Detection" arXiv preprint arXiv:2103.02520