Knowledge graph for multi-source data: optimizing the systematic layout and application of reference materials for edible agricultural products.

In light of the challenges within China's edible agricultural products sector-specifically the fragmented layout, insufficient integration, and underexplored potential of nutrition and safety reference materials-this study aims to better utilize systematic resources. We address the critical lack of structure and weak resource linkage by interlinking diverse data sources, including reference materials, technical testing standards, scientific literature, and research institutions. To establish a robust foundation, we conducted an in-depth analysis of the classification system for edible agricultural products and the framework of nutrition and safety testing indicators. Leveraging visualized knowledge graph technology, we successfully achieved mapping and mutual referencing among these heterogeneous resources. This involved the systematic collection, processing, and integration of massive datasets to construct a comprehensive knowledge base. The developed knowledge graph system has been successfully deployed online, effectively bridging the boundaries between isolated resources. This platform not only facilitates multidimensional knowledge exploration and discovery but also promotes a systematic framework for reference material research. By transforming fragmented data into high-quality, integrated resources, our study significantly enhances the ability to leverage aggregated resources, thereby providing substantial support for the development and application of reference materials in food safety.