The paper introduces a new inference approach for Knowledge Graph (KG) reasoning named Knowledge Pyramid (KP). This approach uses a novel knowledge augmentation strategy to improve the generalization capability of KGs. The framework extracts high-level pyramidal knowledge from low-level knowledge and applies it to reasoning in a multi-level hierarchical KG. The paper shows that the proposed KP model improves the knowledge inference performance with better generalization, especially when there are fewer training samples.
Publication date: 18 Jan 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2401.09070
