Intelligent generation method of emergency plan based on knowledge graph

The traditional emergency response based on static text is usually limited by the experience and ability of the responders, which is low in emergency efficiency and easy to cause omissions. In this study, a knowledge graph framework based on artificial intelligence technology is constructed by integrating multi-source data such as emergency regulation and monitoring. On the basis of this framework, an entity extraction method based on BERT-BILSTM-CRF hybrid model and a relation extraction method based on piecewise convolutional neural network (PCNN) and word segmentation technology are proposed. Then combined with Neo4j graph database, the semantic emergency plan knowledge base is constructed to realize the intelligent storage and visualization of emergency plan. Finally, the intelligent generating page of emergency plan is developed by using computer technology, and the feasibility of the research results is verified, which provides a new technical means for improving the efficiency of emergency response.