Research map / Graph learning / Graph algorithms and clustering
Optimizing Generalized PageRank Methods for Seed-Expansion Community Detection
Pan Li, I Chien, Olgica Milenkovic
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 IPR-d instead of PPR-d or HPR-d | Seed-expansion community detection on synthetic two-block SBMs · Synthetic two-block SBM | recall | Improved |
| Use IPR-d instead of PPR-d or HPR-d | Seed-expansion community detection · Citeseer, Cora, PubMed | recall | Improved |
| Use IPR instead of HPR | Seed-expansion community detection · DBLP | recall | Improved |
| Use IPR instead of HPR | Seed-expansion community detection · Amazon | recall | No clear effect |
| Add degree normalization (ablation) | PageRank methods for seed-expansion community detection · Synthetic SBM, Citeseer, Cora, PubMed | recall | Improved |
| Add degree normalization (ablation) | PageRank methods for seed-expansion community detection · Amazon, DBLP | recall | Hurt |
| Increase the number of walk steps for IPR-d 0.99 (ablation) | Seed-expansion community detection · Citeseer, Cora, PubMed | recall | Improved |
| Use IPR-d with θ=0.95 or θ=0.90 instead of θ=0.99 (ablation) | Seed-expansion community detection · Citeseer, Cora, PubMed | recall | Mixed |
| Use HPR instead of other PR methods at k=5 | Seed-expansion community detection · Amazon, DBLP | recall | Improved |
| Use more walk steps for IPR (ablation) | Seed-expansion community detection · Synthetic two-block SBM | recall | Improved |
Closest papers by meaning
- Adaptive Universal Generalized PageRank Graph Neural Network (ICLR 2021 · similarity 0.88)
- Accelerated Evolving Set Processes for Local PageRank Computation (NeurIPS 2025 · similarity 0.86)
- Query-Limited Community Recovery in Stochastic Block Models (NeurIPS 2026 · similarity 0.86)
- General Graph Random Features (ICLR 2024 · similarity 0.86)
- PageRank Bandits for Link Prediction (NeurIPS 2024 · similarity 0.85)
- Harnessing Multiple Correlated Networks for Exact Community Recovery (NeurIPS 2024 · similarity 0.85)
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.
