Improving Inductive Link Prediction Using Hyper-Relational Facts
This work addresses the challenge of reasoning over unseen entities in knowledge graphs for applications like data integration and AI systems, but it is incremental as it extends existing methods to richer graph structures.
The paper tackles the problem of inductive link prediction on knowledge graphs by using hyper-relational facts, showing that qualifiers over typed edges improve performance by 6% absolute gains in Hits@10 compared to triple-only baselines.
For many years, link prediction on knowledge graphs (KGs) has been a purely transductive task, not allowing for reasoning on unseen entities. Recently, increasing efforts are put into exploring semi- and fully inductive scenarios, enabling inference over unseen and emerging entities. Still, all these approaches only consider triple-based \glspl{kg}, whereas their richer counterparts, hyper-relational KGs (e.g., Wikidata), have not yet been properly studied. In this work, we classify different inductive settings and study the benefits of employing hyper-relational KGs on a wide range of semi- and fully inductive link prediction tasks powered by recent advancements in graph neural networks. Our experiments on a novel set of benchmarks show that qualifiers over typed edges can lead to performance improvements of 6% of absolute gains (for the Hits@10 metric) compared to triple-only baselines. Our code is available at \url{https://github.com/mali-git/hyper_relational_ilp}.