Research map / Graph learning / Knowledge and temporal graphs
Graph Convolutional Networks for Text Classification
Liang Yao, Chengsheng Mao, Yuan Luo
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 a corpus-level graph with document-word and global word-word relations instead of the strongest listed baseline | Text GCN · 20NG test | accuracy: 0.8516 $\pm$ 0.0029 -> 0.8634 $\pm$ 0.0009 | Improved |
| Use a corpus-level graph with document-word and global word-word relations instead of the strongest listed baseline | Text GCN · R8 test | accuracy: 0.9699 $\pm$ 0.0012 -> 0.9707 $\pm$ 0.0010 | Improved |
| Use a corpus-level graph with document-word and global word-word relations instead of the strongest listed baseline | Text GCN · R52 test | accuracy: 0.9320 $\pm$ 0.0004 -> 0.9356 $\pm$ 0.0018 | Improved |
| Use a corpus-level graph with document-word and global word-word relations instead of the strongest listed baseline | Text GCN · Ohsumed test | accuracy: 0.6386 $\pm$ 0.0053 -> 0.6836 $\pm$ 0.0056 | Improved |
| Use a corpus-level graph instead of CNN with pretrained word embeddings | Text GCN · MR test | accuracy: 0.7775 $\pm$ 0.0072 -> 0.7674 $\pm$ 0.0020 | Hurt |
| Use TF-IDF rather than term frequency alone for document-word edge weights (ablation) | Text GCN · Text classification | accuracy | Improved |
| Use PMI rather than word co-occurrence counts for word-word edge weights (ablation) | Text GCN · Text classification | accuracy | Improved |
| Use two GCN layers rather than one (ablation) | Text GCN · Text classification | accuracy | Improved |
| Add more than two GCN layers (ablation) | Text GCN · Text classification | accuracy | No clear effect |
| Increase the sliding-window size from very small values toward 15 (ablation) | Text GCN · R8 and MR | accuracy | Improved |
| Increase the sliding-window size beyond 15 (ablation) | Text GCN · R8 and MR | accuracy | No clear effect |
| Increase the first-layer embedding dimension from too-low values (ablation) | Text GCN · R8 and MR | accuracy | Improved |
Closest papers by meaning
- Continual Graph Convolutional Network for Text Classification (AAAI 2023 · similarity 0.91)
- Tensor Graph Convolutional Networks for Text Classification (AAAI 2020 · similarity 0.90)
- Sparse Structure Learning via Graph Neural Networks for Inductive Document Classification (AAAI 2022 · similarity 0.89)
- Active Discriminative Text Representation Learning (AAAI 2017 · similarity 0.89)
- TGLsta: Low-resource Textual Graph Learning with Semantic and Topological Awareness via LLMs (AAAI 2025 · similarity 0.86)
- Label-free Node Classification on Graphs with Large Language Models (LLMs) (ICLR 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.
