Research map / Graph learning / Graph algorithms and clustering
Active Learning on Attributed Graphs via Graph Cognizant Logistic Regression and Preemptive Query Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Select queries by expected error minimization rather than random selection | SGC active learning · Cora, budget 60 | accuracy: 73.3 -> 79.9 | Helped |
| Select queries by expected error minimization rather than random selection | SGC active learning · Citeseer, budget 60 | accuracy: 70.4 -> 72.8 | Helped |
| Select queries by expected error minimization rather than random selection | SGC active learning · Pubmed, budget 40 | accuracy: 73.9 -> 78.7 | Helped |
| Select queries by expected error minimization rather than random selection | SGC active learning · Am-photo, budget 40 | accuracy: 87.6 -> 90.7* | Helped |
| Select queries by expected error minimization rather than random selection | SGC active learning · Am-comp., budget 40 | accuracy: 73.3 -> 76.8* | Helped |
| Select queries by expected error minimization rather than random selection | SGC active learning · Microwave, budget 60 | accuracy: 83.9 -> 86.0* | Helped |
| Replace sequential risk calculation with preemptive query generation using a predicted previous label (ablation) | GEEM · Cora, budget 60 | accuracy: 79.9 -> 80.3 | No clear effect |
| Replace sequential risk calculation with preemptive query generation using a predicted previous label (ablation) | GEEM · Pubmed, budget 40 | accuracy: 78.7 -> 78.0 | No clear effect |
| Replace sequential risk calculation with preemptive query generation using a predicted previous label (ablation) | GEEM · Am-photo, budget 40 | accuracy: 90.7* -> 89.6 | No clear effect |
| Replace sequential risk calculation with preemptive query generation using a predicted previous label (ablation) | GEEM · Microwave, budget 60 | accuracy: 86.0* -> 86.0* | No clear effect |
| Use the combined model-averaging method instead of label propagation alone | Combined · Transductive node classification, starting with one labelled node | accuracy | Helped |
| Use graph-based logistic regression alongside label propagation for inductive prediction | Combined · Inductive node classification, starting with one labelled node | accuracy | Helped |
Closest papers by meaning
- Combining Label Propagation and Simple Models out-performs Graph Neural Networks (ICLR 2021 · similarity 0.89)
- Label-free Node Classification on Graphs with Large Language Models (LLMs) (ICLR 2024 · similarity 0.89)
- Robust Offline Active Learning on Graphs (NeurIPS 2024 · similarity 0.88)
- Node Identifiers: Compact, Discrete Representations for Efficient Graph Learning (ICLR 2025 · similarity 0.87)
- DGCLUSTER: A Neural Framework for Attributed Graph Clustering via Modularity Maximization (AAAI 2024 · similarity 0.87)
- NLGT: Neighborhood-based and Label-enhanced Graph Transformer Framework for Node Classification (AAAI 2025 · similarity 0.87)
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.