Heterogeneous Topological-Aware Debiasing

Implementation of ECAI 2025 paper "Exploring Topological Bias in Heterogeneous Graph Neural Networks".

Data preparation

Download datasets provided by [1].

[1] Q. Lv, M. Ding, Q. Liu, Y. Chen, W. Feng, S. He, C. Zhou, J. Jiang, Y. Dong, and J. Tang. Are we really making much progress? revisiting, benchmarking and refining heterogeneous graph neural networks. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, pages 1150–1160, 2021.

Investigate topological bias in HGNNs

cd src/scripts
bash run_deg.py

Run the proposed method on three datasets

ACM dataset

cd src/scripts
bash run_acm.sh

IMDB dataset

cd src/scripts
bash run_imdb.sh

DBLP dataset

cd src/scripts
bash run_dblp.sh

Citation

@incollection{zhang2025exploring,
  title={Exploring Topological Bias in Heterogeneous Graph Neural Networks},
  author={Zhang, Yihan},
  booktitle={ECAI 2025},
  pages={2937--2944},
  year={2025},
  publisher={IOS Press}
}