Research map / Learning theory and algorithms / Clustering
Overlapping Clustering Models, and One (class) SVM to Bind Them All
Xueyu Mao, Purnamrita Sarkar, Deepayan Chakrabarti
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 SVM-cone instead of RecoverL2 for topic estimation | NIPS · NIPS semi-synthetic dataset, 20000–60000 documents | ℓ1 reconstruction error: 0.059 -> 0.071; 0.043 -> 0.051; 0.036 -> 0.041 | Hurt |
| Use SVM-cone instead of RecoverL2 for topic estimation | NYT · NYT semi-synthetic dataset, 20000–60000 documents | ℓ1 reconstruction error: 0.125 -> 0.131; 0.103 -> 0.106; 0.095 -> 0.096 | Hurt |
| Use SVM-cone instead of RecoverL2 for topic estimation | PubMed · PubMed semi-synthetic dataset, 20000–60000 documents | ℓ1 reconstruction error: 0.163 -> 0.181; 0.122 -> 0.138; 0.098 -> 0.114 | Hurt |
| Use SVM-cone instead of RecoverL2 for topic estimation | 20NG · 20NG semi-synthetic dataset, 20000–60000 documents | ℓ1 reconstruction error: 0.100 -> 0.090; 0.074 -> 0.064; 0.058 -> 0.052 | Improved |
| Increase the number of documents | SVM-cone topic estimation · NIPS, NYT, PubMed, and 20NG semi-synthetic datasets | ℓ1 reconstruction error | Improved |
| Use SVM-cone instead of competing network baselines | DBLP coauthorship and bipartite author-paper networks · DBLP datasets | average rank correlation | Mixed |
| Use SVM-cone instead of competing topic-model methods | Topic estimation · Semi-synthetic topic datasets | runtime | Improved |
Closest papers by meaning
- Detecting Overlapping and Correlated Communities without Pure Nodes: Identifiability and Algorithm (ICML 2019 · similarity 0.85)
- Non-Exhaustive, Overlapping Clustering (TPAMI 2018 · similarity 0.84)
- Hierarchical Overlapping Clustering on Graphs: Cost Function, Algorithm and Scalability (ICML 2025 · similarity 0.84)
- On Mixed Memberships and Symmetric Nonnegative Matrix Factorizations (ICML 2017 · similarity 0.84)
- Provable Overlapping Community Detection in Weighted Graphs (NeurIPS 2020 · similarity 0.83)
- Recovering Unbalanced Communities in the Stochastic Block Model with Application to Clustering with a Faulty Oracle (NeurIPS 2023 · similarity 0.83)
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.
