he-tiantian / PMP-GNNs

Pytorch implementation of Polarized message-passing graph neural networks published in Artificial Intelligence, 2024.
MIT License
3 stars 1 forks source link
community-detection graph-analytics graph-neural-networks information-network node-classification semi-supervised-learning text-classification

PMP-GNNs

A baseline implementation of Polarized message-passing graph neural networks published in Artificial Intelligence, 2024. The paper is available at: https://www.sciencedirect.com/science/article/pii/S0004370224000651.

Requirements: Python (>=3.8) PyTorch (>=1.9.0) DGL (>=0.9.1)

Please unzip data.zip before running the python files in the corresponding folders.

As there are no pre/post process or early stopping control and different GPU/CUDA platforms might be used, the performances might slightly change.

If you are interested in using the source code and data released here, please cite our paper:

@article{he2024polarized, title={Polarized Message-Passing in Graph Neural Networks}, author={He, Tiantian and Liu, Yang and Ong, Yew-Soon and Wu, Xiaohu and Luo, Xin}, journal={Artificial Intelligence}, pages={104129}, year={2024}, publisher={Elsevier} }