[Purpose/significance]The construction of intelligence analysis knowledge base needs to not only rely on the existing knowledge base construction methods and technologies,but also expand and innovate the key technologies in specific to the information research target. [Method/process] Using the methods of natural language processing and deep learning,this paper explores the text semantic space technology for text knowledge extraction and organization and expert knowledge map technology for expert positioning in the construction of intelligence analysis knowledge base. [Result/conclusion] The paper puts forward the text semantic space including text set,term contribution set and term semantic association set,as well as expert knowledge map including domain terminology association and expert cooperation association. The specific steps are described in detail and experiment is conducted by using graphene field data.