Model-data matching method for natural disaster emergency service scenarios: implementation based on a knowledge graph and community discovery algorithm

Massive disaster-related data and model methods are available in different stages of natural disaster emergency management; however, they often lack the representation of the significance of crucial information. In addition, traditional methods are difficult to automate and fine-match, resulting in greater difficulty in intelligent applications. To address this problem, interconnected knowledge arranged in a knowledge graph and a community discovery algorithm were used to match disaster data with the appropriate method under a unified description framework. Based on the knowledge graph constructed by ontology mapping, this study aimed to represent and store the complex entities and relationships among disaster-related information. The node importance analysis and community discovery algorithm were used to analyze the community structure of important nodes, and a typical rainstorm flood disaster risk assessment emergency service scenario was taken as an example for experimental verification. The results showed that the constructed knowledge graph of natural disaster emergency services can support the formal expression of semantic correlation between disaster scenarios, model methods, and disaster data. The community discovery method performed well in overlapping community module degrees. The method proposed in this study can be used to formalize the representation and storage of massive heterogeneous data and conduct fine matching after quantitative analysis of the importance of model methods and data required for risk assessment of emergency service scenarios. Furthermore, the proposed method can provide theoretical support for improving the scientificity, accuracy, and interpretability of risk assessment decision management and promoting knowledge-driven intelligent emergency information service capabilities.