A method for constructing domain knowledge graph based on literature review
Shulin Hu, Shuxuan Wang, Huajun Zhang, Xuesong Hu · 2022
Domain knowledge graph, as a branch of knowledge graph, focuses both the coverage and usage of knowledge in a specific field, so it has high requirements for the depth and accuracy of knowledge in this field. At present, there are still problems in the rapid construction and application of domain knowledge graphs. Especially before construction, complex domain scenarios involve a wide range of knowledge dimensions, and it is difficult for business experts to comprehensively construct a conceptual data model in this domain in a short period of time. In response to this problem, this paper proposes a domain knowledge graph construction method based on review documents, extracts the concepts and definitions of related fields in the review documents, constructs a domain knowledge graph, and uses NEO4J to visualize the knowledge graph from the extracted knowledge data set. In this paper, the field of spatiotemporal prediction is used as the experimental object to construct a knowledge graph in this field. The results show that this method can construct a knowledge graph framework with high accuracy in a short time. The domain knowledge graph is small and has yet to be expanded.