Type-aware Embeddings for Multi-Hop Reasoning over Knowledge Graphs

TL;DR

This paper introduces TEMP, a type-aware message passing model that enhances multi-hop reasoning over knowledge graphs by incorporating type information, leading to improved reasoning accuracy and generalization.

Contribution

The paper proposes a novel TEMP model that leverages type information in knowledge graphs, improving multi-hop reasoning and can be integrated into existing embedding methods.

Findings

TEMP outperforms existing models on three real-world datasets.

Incorporating type information improves reasoning accuracy.

TEMP enhances generalization and inductive reasoning capabilities.

Abstract

Multi-hop reasoning over real-life knowledge graphs (KGs) is a highly challenging problem as traditional subgraph matching methods are not capable to deal with noise and missing information. To address this problem, it has been recently introduced a promising approach based on jointly embedding logical queries and KGs into a low-dimensional space to identify answer entities. However, existing proposals ignore critical semantic knowledge inherently available in KGs, such as type information. To leverage type information, we propose a novel TypE-aware Message Passing (TEMP) model, which enhances the entity and relation representations in queries, and simultaneously improves generalization, deductive and inductive reasoning. Remarkably, TEMP is a plug-and-play model that can be easily incorporated into existing embedding-based models to improve their performance. Extensive experiments on three real-world datasets demonstrate TEMP's effectiveness.