Research map / Graph learning / Knowledge and temporal graphs
Inductive Meta-Path Learning for Schema-Complex Heterogeneous Information Networks
Shixuan Liu, Changjun Fan, Kewei Cheng, Yunfei Wang, Peng Cui, Yizhou Sun, Zhong 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 |
|---|---|---|---|
| increase random walk search attempts to 5 times | Random Walk · YAGO26K-906 inductive | MRR: 0.089 -> 0.151 | Improved |
| increase random walk search attempts to 10 times | Random Walk · YAGO26K-906 inductive | MRR: 0.089 -> 0.174 | Improved |
| replace random walk with SchemaWalk at equal attempts | SchemaWalk · DB111K-174 inductive | MRR: 0.012 -> 0.731 | Improved |
| replace exhaustive enumeration with SchemaWalk | SchemaWalk · YAGO26K-906 inductive | MRR: 0.400 -> 0.218 | Hurt |
| move from transductive to inductive multi-relation setting | SchemaWalk · YAGO26K-906/DB111K-174 | Hits/MRR: average decline of 3.32% | No clear effect |
| remove 100% instead of 50% of entity nodes | SchemaWalk · NELL PlaysAgainst entity-level inductive | ROC-AUC / AP: ROC-AUC dropped by 15.1%, AP by 6% | Hurt |
| set lambda_1 excessively large (ablation) | SchemaWalk · DB111K-174 MRR / YAGO isCitizenOf AUC | MRR / AUC: deteriorates by 12% and 7% | Hurt |
| use SchemaWalk instead of RotatE | SchemaWalk · NELL CompetesWith | ROC-AUC: 0.870 -> 0.706 | Hurt |
| use SchemaWalk instead of RotatE | SchemaWalk · YAGO26K-906 DiedIn | ROC-AUC: 0.860 -> 0.947 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- RotatE — RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space (ICLR 2019) · better on 2 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Reinforcement Learning Based Meta-Path Discovery in Large-Scale Heterogeneous Information Networks (AAAI 2020 · similarity 0.93)
- Heterogeneous Graph Neural Network on Semantic Tree (AAAI 2025 · similarity 0.85)
- Anchoring Path for Inductive Relation Prediction in Knowledge Graphs (AAAI 2024 · similarity 0.85)
- Replacing Paths with Connection-Biased Attention for Knowledge Graph Completion (AAAI 2025 · similarity 0.84)
- Towards Global-Topology Relation Graph for Inductive Knowledge Graph Completion (AAAI 2025 · similarity 0.83)
- Harnessing Language Model for Cross-Heterogeneity Graph Knowledge Transfer (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.
