Link Prediction on N-ary Relational Facts: A Graph-based Approach
arXiv:2105.08476
Abstract
Link prediction on knowledge graphs (KGs) is a key research topic. Previous work mainly focused on binary relations, paying less attention to higher-arity relations although they are ubiquitous in real-world KGs. This paper considers link prediction upon n-ary relational facts and proposes a graph-based approach to this task. The key to our approach is to represent the n-ary structure of a fact as a small heterogeneous graph, and model this graph with edge-biased fully-connected attention. The fully-connected attention captures universal inter-vertex interactions, while with edge-aware attentive biases to particularly encode the graph structure and its heterogeneity. In this fashion, our approach fully models global and local dependencies in each n-ary fact, and hence can more effectively capture associations therein. Extensive evaluation verifies the effectiveness and superiority of our approach. It performs substantially and consistently better than current state-of-the-art across a variety of n-ary relational benchmarks. Our code is publicly available.
arXiv commentAccepted to Findings of ACL 2021
References in corpus (8)
- Gaussian Error Linear Units (GELUs)
- A Survey on Knowledge Graphs: Representation, Acquisition and Applications
- A Review of Relational Machine Learning for Knowledge Graphs
- RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
- CoKE: Contextualized Knowledge Graph Embedding
- Improving Hyper-Relational Knowledge Graph Completion
- Entity Structure Within and Throughout: Modeling Mention Dependencies for Document-Level Relation Extraction
- Role-Aware Modeling for N-ary Relational Knowledge Bases