Research on construction method of knowledge graph in the civil aviation security field

The knowledge graph of the civil aviation safety domain can visually display the complex network relationship between civil aviation safety-related factors, and facilitate effective analysis of potential security risks from the relationship between complex and multi-dimensional security factors, providing in-depth risk analysis and decision support to the field. In this paper, the knowledge system of the civil aviation safety field was firstly combed, and the ontology model suitable for the civil aviation security field has been constructed. Then knowledge was extracted from structured, semi-structured and unstructured data sources in the civil aviation security field. On this basis, entity recognition of unstructured text has been done by using a machine-learning correlation model algorithm, and the entity-relationship in the civil aviation security field has been extracted by using a deep convolutional neural network model combined with keyword features. Finally, the map database Neo4j was used to store the knowledge of the civil aviation safety field, and the knowledge map of the civil aviation safety field has been initially constructed.