Research map / Graph learning / Knowledge and temporal graphs
Deep Temporal Graph Clustering: A Comprehensive Benchmark and Datasets
Meng Liu, Ke Liang, Siwei Wang, Xingchen Hu, Sihang Zhou, Xinwang Liu
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 |
|---|---|---|---|
| apply BenchTGC framework clustering modules | HTNE · arXivCS | ACC: 25.57 -> 41.49 | Improved |
| apply BenchTGC framework clustering modules | HTNE · arXivAI | ACC: 65.66 -> 76.06 | Improved |
| apply BenchTGC framework clustering modules | TREND · DBLP | ACC: 46.82 -> 55.10 | Improved |
| apply BenchTGC framework clustering modules | S2T · Brain | ACC: 34.32 -> 48.86 | Improved |
| apply BenchTGC framework across 5 temporal methods and 9 datasets | temporal graph methods · all 9 datasets | ACC/NMI/ARI/F1: only 2 comparisons showed no improvement out of 180 | Improved |
| add BenchTGC modules to JODIE (cross-batch calibration) | JODIE · all datasets | clustering | No clear effect |
| add cluster-center modules to MNCI | MNCI · all datasets | clustering | No clear effect |
| use temporal interaction sequence instead of static adjacency matrix | IJODIE/ITREND vs static methods · School | ACC/NMI/ARI/F1: 48.32 -> 100.0 | Improved |
| use static adjacency-based clustering on large graphs instead of batched temporal training | static clustering methods · arXivCS and larger | GPU memory: 16.19 GB vs 8.93 GB | Hurt |
| reduce batch size to 1 (ablation) | temporal methods · Brain | runtime | Hurt |
| reduce batch size to 1 (ablation) | temporal methods · School/Brain/Patent/arXivAI | GPU memory: below 0.5 GB | Improved |
| increase batch size from 128 to 10000 (ablation) | temporal methods · School/Brain/Patent/arXivAI | runtime: less than 5 minutes per epoch | Improved |
Closest papers by meaning
- Deep Temporal Graph Clustering (ICLR 2024 · similarity 0.95)
- TGB 2.0: A Benchmark for Learning on Temporal Knowledge Graphs and Heterogeneous Graphs (NeurIPS 2024 · similarity 0.89)
- DGCLUSTER: A Neural Framework for Attributed Graph Clustering via Modularity Maximization (AAAI 2024 · similarity 0.87)
- State Space Models on Temporal Graphs: A First-Principles Study (NeurIPS 2024 · similarity 0.87)
- Learning Persistent Community Structures in Dynamic Networks via Topological Data Analysis (AAAI 2024 · similarity 0.86)
- Every Node Is Different: Dynamically Fusing Self-Supervised Tasks for Attributed Graph Clustering (AAAI 2024 · 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.
