A General Solution and Practice for Automatically Constructing Domain Knowledge Graph

Abstract

A knowledge graph is a semantic network that reveals the relationships between entities. Domain knowledge graphs can assist practitioners in complex analysis applications and decision support, and build industry barriers. A good construction system can help companies construct domain knowledge graphs efficiently and quickly. However, due to the differences in data models and business requirements in different fields, there is no general method for constructing a domain knowledge graph. Moreover, the current construction methods require a lot of manpower investment, which is relatively inefficient. To solve the inefficiency and nonuniversal problems in the process of domain knowledge graph construction, in this paper, we study the automated construction mechanism of Chinese domain knowledge graphs and propose a general automatic construction solution. Experimental results show that our proposed solution is more effective in constructing a domain knowledge graph.