Research on Supply Chain Knowledge Graph Inference Method Based on Quaternion Embedding

Supply chain reasoning refers to the analysis and inference of supply chain-related data to understand the behavior of potential entities and relationships within the supply chain. By fully exploring latent information and complementing the existing supply chain, it can effectively enhance the resilience of the supply chain and its ability to respond to risks. Traditional research mainly relies on supply chain networks for reasoning, relying on the structural properties of the network itself, and fails to fully utilize the knowledge information of nodes and relationships. In this study, knowledge graphs are applied to supply chain reasoning, and an improved quaternion model is used f or supply chain reasoning. The proposed method is validated u sing an example of an automotive supply chain. The research results show that compared to traditional supply chain reasoning methods, the proposed method can fully utilize the advantages of knowledge information in the supply chain and the co-occurrence properties of nodes in the knowledge graph, thereby imp roving the reasoning effectiveness.