Abstract: 【Purpose/significance】The research makes a systematic study of the knowledge flow pattern in the citation network with
the method of network analysis, so as to provide theoretical and empirical basis for the knowledge diffusion, transformation and innova-tion in the citation network.【Method/process】Descriptive statistics and network analysis indexes is selected to describe and analyze the ability and role of knowledge flow of knowledge nodes, the type and structure of knowledge flow of knowledge community, the distri-bution and structure characteristics of knowledge flow of the whole network.【Result/conclusion】 According to CNKI database, cita-tion networks of "Think Tank", "Digital Humanities" and "Data Governance" is constructed respectively, and the similarities and dif-ferences of knowledge flow characteristics between them is compared and analyzed according to the methods in this paper. The method in this paper can deeply mine the knowledge association among academic documents, and make up for the defects of neglecting the deep-seated reference information in the past research.【Innovation/limitation】This article uses a variety of indicators and methods to systematically study the laws and patterns of citation network knowledge flow from the perspective of citation content, but does not ex-tract individual citation networks that reflect one or some knowledge attributes from the overall citation network for analysis .