Hierarchy-Aware Semantic Losses for Knowledge Graph Link Prediction
2026-08-24 • Machine Learning
Machine Learning
AI summaryⓘ
The authors studied how to improve predicting missing links in knowledge graphs by using information from class hierarchies, which organize concepts in a parent-child way. They tested a method that encourages the model to respect these hierarchies using special 'semantic losses' during training, instead of just adding extra connections in the graph. Their experiments showed this method worked better across several datasets than both standard models and those that simply add hierarchy edges. This suggests that teaching models to understand hierarchical relationships explicitly helps them make more accurate predictions in knowledge graphs.
Knowledge GraphOntologyClass HierarchyGraph Neural Network (GNN)Link PredictionBox EmbeddingsSemantic LossMean Reciprocal Rank (MRR)Subclass Relations
Authors
Filip Kronström, Ross D. King
Abstract
Knowledge graphs are often accompanied by ontological class hierarchies that encode valuable semantic information, yet many link prediction methods either ignore such hierarchies or incorporate them indirectly through additional graph edges. Recent work introduced hierarchy-aware graph neural networks (GNNs), which use semantic losses derived from box embeddings to encourage satisfaction of subclass relationships during GNN-based representation learning. While this approach has shown promise for biological regression tasks, its effectiveness for knowledge graph link prediction has not been investigated. In this paper we evaluate hierarchy-aware semantic losses on link prediction across three benchmark datasets: AIFB, CoDEx, and BioKG. We combine graph neural network encoders with box-embedding-based semantic losses that encourage learned representations to better satisfy ontology-derived class hierarchies, and compare this approach to both standard link prediction models and models incorporating subclass relations as graph edges. Across all datasets, hierarchy-aware semantic losses significantly improve mean reciprocal rank (MRR) and consistently outperform models that incorporate hierarchy information through additional subclass edges. Relative to the baseline GNN models, MRR improved by 7.6%, 2.4%, and 15.5% on AIFB, CoDEx, and BioKG, respectively. Furthermore, semantic losses consistently outperform the alternative of augmenting the graph with subclass edges. These results are consistent with ontology-derived class hierarchies providing complementary information to graph structure, and suggest that encouraging hierarchical consistency through semantic losses is an effective and comparatively parameter-efficient mechanism for improving knowledge graph link prediction.