Intelligent sound image archive editing and research method based on knowledge graph

The invention relates to the technical field of electronic archive architecture, and discloses an intelligent sound image archive editing and research method based on a knowledge graph, and the implementation method comprises the following steps: extracting structured data in a sound image system archive into a triple to construct the knowledge graph, and then receiving user topic selection information, and for an entity lacking user topic selection information, automatically generating an editing and research target label by using the sound image file intelligent editing and research model, storing the editing and research target label in a knowledge graph database, and retrieving related sound image data from the knowledge graph for editing and research. Related features in the audio and the picture are automatically extracted according to the editing and research targets, so that the sound image entity is represented by different embedded vectors under different editing and research targets, and then editi