Research map / Graph learning / Knowledge and temporal graphs
Improved Knowledge Graph Embedding Using Background Taxonomic Information
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 |
|---|---|---|---|
| Add non-negative entity embeddings without incorporating subsumptions (ablation) | SimplE · WN18 | filtered MRR: 0.942 -> 0.937 | No clear effect |
| Add non-negative entity embeddings without incorporating subsumptions (ablation) | SimplE · FB15k | filtered MRR: 0.727 -> 0.725 | No clear effect |
| Use ReLU rather than exponential or logistic to produce non-negative embeddings (ablation) | SimplE+ · WN18 and FB15k | filtered MRR: WN18: 0.866 -> 0.937; 0.854 -> 0.937; FB15k: 0.575 -> 0.725; 0.425 -> 0.725 | Helped |
| Add taxonomic subsumption information to the embedding method | SimplE+ · Sport | Hit@1: 0.184 -> 0.349 | Helped |
| Add taxonomic subsumption information to the embedding method | SimplE+ · Location | Hit@1: 0.130 -> 0.430 | Helped |
| Combine learned embeddings with taxonomic inference rather than using either alone | SimplE+ · Sport, across tested training-data fractions | Hit@1 | Helped |
| Use taxonomic information during training | SimplE+ · Sport | epochs to convergence | Helped |
Closest papers by meaning
- LENA: Locality-Expanded Neural Embedding for Knowledge Base Completion (AAAI 2019 · similarity 0.90)
- Learning Triple Embeddings from Knowledge Graphs (AAAI 2020 · similarity 0.90)
- SimplE Embedding for Link Prediction in Knowledge Graphs (NeurIPS 2018 · similarity 0.89)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.88)
- Toward a Unified Statistical Theory of Unsupervised Pretraining and Supervised Neural Knowledge Graph Learning (NeurIPS 2026 · similarity 0.88)
- Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization (ICML 2024 · 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.