This paper combines the knowledge related to the Knowledge Graph to store and manage the ideological knowledge in the schema layer in the form of an ontology repository. The Bootstrapping-based extraction method has been improved for entity relationship extraction at the data layer, and semantically constrained relationships have been added to reduce the problem of extraction bias. Civics education’s characteristic attributes are identified through attribute extraction using crowdsourcing. For the entity alignment problem in entity extraction, the relationship triad and attribute triad are jointly used for entity vector characterization, and the entity alignment model based on BERT and TransE is proposed. Finally, in the process of testing knowledge mapping, more than 70% of college students were satisfied, and the percentage of college students agreeing and strongly agreeing in the dimensions of knowledge construction were 68.81%, 77.92%, and 75.63%, respectively. In the six dimensions of positive psychological quality of college students who have used knowledge mapping, there is a significant increase in the data of the courage dimension from 3.4 to 4.14, and the other dimensions have also increased. The performance of the knowledge graph constructed in this paper for civic information education is outstanding.