Research map / Graph learning / Knowledge and temporal graphs
Graph-Based Wrong IsA Relation Detection in a Large-Scale Lexical Taxonomy
Jiaqing Liang, Yanghua Xiao, Yi Zhang, Seung-won Hwang, Haixun Wang
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 hyponym-number factor Ph to frequency weight (wfh vs wf) (ablation) | MFAS · Probase wrong isA detection | precision (%): 86.0% -> 90.7% | Improved |
| add hyponym-number factor Ph to frequency weight (wfh vs wf) (ablation) | Agony+ · Probase wrong isA detection | precision (%): 85.7% -> 91.3% | Improved |
| add hyponym-number factor Ph to frequency weight (wfh vs wf) (ablation) | Agony · Probase wrong isA detection | precision (%): 83.7% -> 84.7% | Improved |
| use Agony+ edge-removal optimization instead of removing all backward edges (ablation) | Agony#2 · Probase wrong isA detection | precision (%): 84.7% -> 91.3% | Improved |
| use Agony+ optimization with wf weight (ablation) | Agony#1 · Probase wrong isA detection | relative recall: 37.5% -> 23.6% | Hurt |
| replace topological-sort baseline with graph-based MFAS/Agony models | cycle-based wrong isA detection · Probase wrong isA detection | precision (%): 71.0% -> 91.3% | Improved |
| use optimized backward-edge removal in topological baseline (Baseline+) (ablation) | Baseline · Probase wrong isA detection | precision (%): 71.0% -> 72.3% | No clear effect |
| use Agony#2 level assignment instead of topological baseline | level assignment · Probase level assignment vs WordNet | Pearson r: 0.568 -> 0.692 | Improved |
Closest papers by meaning
- Reasoning of Large Language Models over Knowledge Graphs with Super-Relations (ICLR 2025 · similarity 0.85)
- Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (ICLR 2024 · similarity 0.83)
- Context-aware Inductive Knowledge Graph Completion with Latent Type Constraints and Subgraph Reasoning (AAAI 2025 · similarity 0.83)
- Text2NKG: Fine-Grained N-ary Relation Extraction for N-ary relational Knowledge Graph Construction (NeurIPS 2024 · similarity 0.83)
- Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs (NeurIPS 2025 · similarity 0.83)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (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.
