Research map / Graph learning / GNN architectures
Characterizing Graph Datasets for Node Classification: Homophily-Heterophily Dichotomy and Beyond
Oleg Platonov, Denis Kuznedelev, Artem Babenko, Liudmila Prokhorenkova
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use label informativeness (LI) instead of adjusted homophily to explain GNN accuracy | GraphSAGE · synthetic SBM graphs | Spearman correlation with accuracy: 0.05 -> 0.93 | Improved |
| use LI instead of adjusted homophily to explain GNN accuracy | GCN · synthetic SBM graphs | Spearman correlation with accuracy: 0.19 -> 0.76 | Improved |
| use LI instead of adjusted homophily to explain GNN accuracy | GAT · synthetic SBM graphs | Spearman correlation with accuracy: 0.17 -> 0.77 | Improved |
| use LI instead of homophily measures to explain GNN accuracy | GCN · semi-synthetic cora | Spearman correlation with accuracy: -0.31 -> 0.72 | Improved |
| use LI instead of homophily measures to explain GNN accuracy | GraphSAGE · semi-synthetic citeseer | Spearman correlation with accuracy: -0.54 -> 0.51 | Improved |
| use adjusted homophily instead of edge homophily to characterize binary datasets | questions dataset · questions | homophily value: 0.84 -> 0.02 | Mixed |
Closest papers by meaning
- What Is Missing For Graph Homophily? Disentangling Graph Homophily For Graph Neural Networks (NeurIPS 2024 · similarity 0.94)
- Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs (NeurIPS 2020 · similarity 0.93)
- On the Impact of Feature Heterophily on Link Prediction with Graph Neural Networks (NeurIPS 2024 · similarity 0.92)
- Understanding Heterophily for Graph Neural Networks (ICML 2024 · similarity 0.90)
- PC-Conv: Unifying Homophily and Heterophily with Two-Fold Filtering (AAAI 2024 · similarity 0.90)
- HeroFilter: Adaptive Spectral Graph Filter for Varying Heterophilic Relations (NeurIPS 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
