Research map / Learning theory and algorithms / Clustering
Spectral Algorithms for Community Detection in Directed Networks
Zhe Wang, Yingbin Liang, Pengsheng Ji
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 |
|---|---|---|---|
| add ratio step (D-SCORE) to remove heterogeneous degree parameters (ablation) | oPCA · Political blogs, entire graph | misclustered nodes: 434 -> 142 | Improved |
| add ratio step to regularized Laplacian version (rD-SCORE) (ablation) | rPCA · Political blogs, entire graph | misclustered nodes: 414 -> 139 | Improved |
| add intersection-with-attachment (ablation) | D-SCORE · Political blogs | misclustered nodes: 142 -> 60 | Improved |
| add intersection-with-attachment (ablation) | oPCA · Political blogs | misclustered nodes: 434 -> 300 | Improved |
| add intersection-with-attachment (ablation) | D-SCORE2 · Email-Eu-core | misclustered nodes: 15 -> 4 | Improved |
| add intersection-with-attachment (ablation) | D-SCORE · Email-Eu-core | misclustered nodes: 23 -> 7 | Improved |
| add regularized graph Laplacian preprocessing (ablation) | D-SCORE · Political blogs, entire graph | misclustered nodes: 142 -> 139 | No clear effect |
| replace l2 row normalization (D-SCORE2) by the default D-SCORE (ablation) | D-SCORE · Political blogs, entire graph | misclustered nodes: 141 vs 142 | No clear effect |
| use D-SCORE instead of APL | directed community detection · Political blogs, entire graph | misclustered nodes: 61 vs 60 (with attachment) | Mixed |
| use D-SCORE instead of BCPL | directed community detection · Political blogs, entire graph | misclustered nodes: 379 -> 142 | Improved |
| use intersection-with-attachment on dense asymmetric DCBM (ablation) | D-SCORE variants · Simulation: DCBM asymmetric and dense | misclustering rate | No clear effect |
| use intersection-with-attachment on sparse DCBM (ablation) | D-SCORE variants · Simulation: DCBM asymmetric and sparse | misclustering rate | Improved |
Closest papers by meaning
- Analysis of spectral clustering algorithms for community detection: the general bipartite setting (JMLR 2019 · similarity 0.89)
- Community models for networks observed through edge nominations (JMLR 2023 · similarity 0.89)
- Semi-supervised Community Detection via Structural Similarity Metrics (ICLR 2023 · similarity 0.88)
- Improved spectral community detection in large heterogeneous networks (JMLR 2018 · similarity 0.88)
- Determining the Number of Communities in Degree-corrected Stochastic Block Models (JMLR 2021 · similarity 0.87)
- Optimal Graph Clustering without Edge Density Signals (NeurIPS 2025 · similarity 0.86)
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.
