Research map / Graph learning / Graph algorithms and clustering
An Optimization-based Approach To Node Role Discovery in Networks: Approximating Equitable Partitions
Michael Scholkemper, Michael T Schaub
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 the EV-based graph embedding instead of Graph2Vec | Few-shot graph classification · AIDS | accuracy: 79.9± 4.4 -> 95.0 ± 5.2 | Improved |
| Use the EV-based graph embedding instead of Graph2Vec | Few-shot graph classification · ENZYMES | accuracy: 20.3± 1.5 -> 21.3± 1.7 | Improved |
| Use the EV-based graph embedding instead of Graph2Vec | Few-shot graph classification · PROTEINS | accuracy: 60.3± 3.5 -> 66.5± 6.4 | Improved |
| Use the EV-based graph embedding instead of Graph2Vec | Few-shot graph classification · NCI1 | accuracy: 53.9± 1.5 -> 58.5± 4.0 | Improved |
| Use the EV-based graph embedding instead of Graph2Vec | Few-shot graph classification · MUTAG | accuracy: 66.9± 5.6 -> 81.5± 7.7 | Improved |
| Train GIN on 90% of the dataset instead of using the EV embedding with 10 training examples | Graph classification · AIDS | accuracy: 95.0 ± 5.2 -> 97.8 ± 1.4 | Improved |
| Train GIN on 90% of the dataset instead of using the EV embedding with 10 training examples | Graph classification · ENZYMES | accuracy: 21.3± 1.7 -> 60.3 ± 0.7 | Improved |
| Train GIN on 90% of the dataset instead of using the EV embedding with 10 training examples | Graph classification · PROTEINS | accuracy: 66.5± 6.4 -> 75.4 ± 1.3 | Improved |
| Train GIN on 90% of the dataset instead of using the EV embedding with 10 training examples | Graph classification · NCI1 | accuracy: 58.5± 4.0 -> 82.0 ± 0.3 | Improved |
| Train GIN on 90% of the dataset instead of using the EV embedding with 10 training examples | Graph classification · MUTAG | accuracy: 81.5± 7.7 -> 94.3 ± 0.5 | Improved |
| Use fuzzy c-means rather than average linkage in approximate WL | Approximate WL on RIP planted-role recovery · RIP planted-role recovery | recovery accuracy | Improved |
| Use EV-based clustering rather than the other tested methods | Long-term cost minimization · RIP planted-role recovery | long-term cost | Improved |
Closest papers by meaning
- Community models for networks observed through edge nominations (JMLR 2023 · similarity 0.85)
- Crawling the Community Structure of Multiplex Networks (AAAI 2019 · similarity 0.84)
- An Efficient Local Search Approach for Polarized Community Discovery in Signed Networks (NeurIPS 2025 · similarity 0.84)
- Learning Persistent Community Structures in Dynamic Networks via Topological Data Analysis (AAAI 2024 · similarity 0.84)
- Optimal Quasi-clique: Hardness, Equivalence with Densest-k-Subgraph, and Quasi-partitioned Community Mining (AAAI 2024 · similarity 0.83)
- Solving Multiagent Path Finding on Highly Centralized Networks (AAAI 2025 · 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.
