Knowledge Graph Embeddings with Representing Relations as Annular Sectors

2025-06-06

AI Summary

Knowledge graph completion (KGC) struggles with modeling the semantic hierarchical structure of entities, particularly in region-based embedding models where entities are typically represented as points and relations as geometric regions—limiting their capacity to capture hierarchical semantics. To address this, we propose SectorE, the first KGC model that represents relations as annular sectors in polar coordinates. In SectorE, entity magnitudes encode hierarchical depth, while phases encode semantic roles, enabling entities to be naturally embedded within relation-specific sectors. This design preserves geometric reasoning capabilities while explicitly modeling semantic hierarchies. SectorE achieves state-of-the-art or highly competitive performance on FB15k-237, WN18RR, and YAGO3-10, demonstrating significant improvements in modeling and predicting hierarchical relations.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Knowledge graphs (KGs), structured as multi-relational data of entities and relations, are vital for tasks like data analysis and recommendation systems. Knowledge graph completion (KGC), or link prediction, addresses incompleteness of KGs by inferring missing triples (h, r, t). It is vital for downstream applications. Region-based embedding models usually embed entities as points and relations as geometric regions to accomplish the task. Despite progress, these models often overlook semantic hierarchies inherent in entities. To solve this problem, we propose SectorE, a novel embedding model in polar coordinates. Relations are modeled as annular sectors, combining modulus and phase to capture inference patterns and relation attributes. Entities are embedded as points within these sectors, intuitively encoding hierarchical structure. Evaluated on FB15k-237, WN18RR, and YAGO3-10, SectorE achieves competitive performance against various kinds of models, demonstrating strengths in semantic modeling capability.

Problem

• Address incompleteness in knowledge graphs via link prediction
• Capture semantic hierarchies in entities and relations
• Model relations as annular sectors for better representation

Innovation

• Embeds relations as annular sectors
• Uses polar coordinates for modeling
• Encodes hierarchical structure intuitively