(College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China)
Abstract： With the rapid development of the industrial Internet, the demand for natural language processing applications such as building knowledge graphs is gradually increasing in the face of massive industrial data. The effective management and mining of industrial information can help to discover the risks and opportunities faced in time, and the risk prediction of the industrial chain can provide regulatory authorities with early warning means for industrial risks. In view of the above problems, this paper takes the knowledge related to knowledge graph as the scientific basis, and puts forward the criteria for labeling industrial text data entities based on knowledge graph technology, extracts knowledge from massive listed companies′ industrial information, and forms a topdown threedimensional industrial knowledge map. At the same time, the intrinsic relationship of specific industrial chain knowledge of listed enterprises in the industrial knowledge graph is studied, the law is summarized, and the graph reasoning is realized by combining the characteristic information of the time series chart of the industrial chain in previous years, and the market value of listed enterprises in the industrial chain is successfully predicted and analyzed
Key words : knowledge graph; industry chain analysis; risk prediction; entity relationship callouts
