Embedding-Based Asynchronous Entity Classification Algorithm Framework for the Defect Knowledge Graph of Distribution Network Equipment

The defect record of distribution network equipment can form a fault report and provide data support for related users. Knowledge graph can be used to realize the knowledge interconnection of distribution network equipment defect records. Entity classification is a very important sub-task in the complete task of knowledge graph, which is beneficial to the perfection of skeleton structure in knowledge graph. Therefore, it is of great significance to study the entity classification technology of distribution network equipment defects. At present, the entity classification technology of knowledge graphs can be divided into two types: asynchronous-based entity classification method and synchronous-based entity classification method. This paper proposed a new embedding-based asynchronous entity classification algorithm framework. Compared with entity classification based on the synchronous training method, the performance of the proposed method was verified.