Construction of a Knowledge Graph for the Principles of Communication Course Based on System Thinking

From the perspective of a complete communication system, this paper presents the construction of a knowledge graph for the Principles of Communication course through a semi-automatic method combining automated techniques with human verification. The ontology is designed with a hierarchical structure of "complete system - various transmission system types - knowledge points." Entity recognition is performed using a Bi-LSTM+CRF model combined with manual verification, followed by relation extraction using a Relational Graph Convolutional Network (R-GCN). After attribute extraction and knowledge fusion, visualization is achieved using Neo4j. The resulting knowledge graph integrates the structure of the communication system, enabling students to intuitively understand the relationships between knowledge points, plan learning paths, and enhance learning efficiency.