Substructure-aware subgraph reasoning for inductive relation prediction

Relation prediction aims to infer the missing relations among entities in knowledge graphs, where inductive relation prediction enjoys great popularity due to its effectiveness to be applied to emerging entities. Most existing approaches learn the logical compositional rules or utilize subgraphs to predict the missing relation. Although great progress has been made in the performance, current models are still suboptimal due to their limited ability to capture topological information that is critical for local relation prediction. To address this problem, we propose a novel inductive relation prediction approach called substructure-aware subgraph reasoning which incorporates the substructure information of subgraphs into the reasoning process, thus making the relation prediction more precise. Specifically, we extract the entities and relations around the target entities to form the subgraph and then encode the structure information of nodes and edges by counting the number of certain substructures. Next, the structural information is explicitly applied to the message passing for more accurate reasoning. To improve the performance, we also utilize the semantic correlations between relations as auxiliary information. Experimental results on three benchmark datasets show the effectiveness of the proposed approach for the inductive relation prediction.

Automated summary

Relation prediction aims to infer missing relations among entities in knowledge graphs, and inductive relation prediction has gained popularity for its effectiveness on emerging entities. Existing approaches either learn logical compositional rules or utilize subgraphs for relation prediction. However, current models are still suboptimal in capturing critical topological information for local relation prediction. To address this problem, we propose a novel inductive relation prediction approach called substructure-aware subgraph reasoning, which incorporates substructure information into the reasoning process to improve prediction accuracy. Experimental results on three benchmark datasets demonstrate the effectiveness of our proposed approach for inductive relation prediction.