Research map / Graph learning / GNN architectures
Tensor Graph Convolutional Networks for Text Classification
Xien Liu, Xinxin You, Xiao Zhang, Ji Wu, Ping Lv
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 SemGraph to the graph tensor (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8688 -> 0.8774; 0.9776 -> 0.9804; 0.9334 -> 0.9505; 0.6917 -> 0.7011; 0.7763 -> 0.7791 | Mixed |
| Add SynGraph to the graph tensor (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8707 -> 0.8774; 0.9753 -> 0.9804; 0.9529 -> 0.9505; 0.6854 -> 0.7011; 0.7710 -> 0.7791 | Mixed |
| Add SeqGraph to the graph tensor (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8699 -> 0.8774; 0.9753 -> 0.9804; 0.9486 -> 0.9505; 0.6921 -> 0.7011; 0.7746 -> 0.7791 | Mixed |
| Add inter-graph propagation after intra-graph propagation (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8701 -> 0.8774; 0.9808 -> 0.9804; 0.9498 -> 0.9505; 0.6723 -> 0.7011; 0.7656 -> 0.7791 | Mixed |
| Preserve separate graph representations and learn with TensorGCN instead of merging edges (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8721 -> 0.8774; 0.9749 -> 0.9804; 0.9463 -> 0.9505; 0.6951 -> 0.7011; 0.7532 -> 0.7791 | Mixed |
| Use only SeqGraph rather than all three graphs (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8634 -> 0.8774; 0.9707 -> 0.9804; 0.9356 -> 0.9505; 0.6836 -> 0.7011; 0.7674 -> 0.7791 | Hurt |
| Use only SynGraph rather than all three graphs (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8478 -> 0.8774; 0.9534 -> 0.9804; 0.9299 -> 0.9505; 0.6658 -> 0.7011; 0.7496 -> 0.7791 | Hurt |
| Use only SemGraph rather than all three graphs (ablation) | TensorGCN · 20NG, R8, R52, Ohsumed, MR | accuracy: 0.8517 -> 0.8774; 0.9520 -> 0.9804; 0.9245 -> 0.9505; 0.6626 -> 0.7011; 0.7620 -> 0.7791 | Hurt |
Closest papers by meaning
- Graph Convolutional Networks for Text Classification (AAAI 2019 · similarity 0.90)
- Semi-Supervised Classification with Graph Convolutional Networks (ICLR 2017 · similarity 0.88)
- An End-to-End Deep Learning Architecture for Graph Classification (AAAI 2018 · similarity 0.87)
- Graph Classification via Reference Distribution Learning: Theory and Practice (NeurIPS 2024 · similarity 0.86)
- Bootstrapping Heterogeneous Graph Representation Learning via Large Language Models: A Generalized Approach (AAAI 2025 · similarity 0.86)
- TGLsta: Low-resource Textual Graph Learning with Semantic and Topological Awareness via LLMs (AAAI 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.
