Abstract: Traditional Chinese medicine classics encompass the foundational theories, pharmacological knowledge, and practical experience of Chinese medicine over thousands of years, possessing significant research value. However, manual reading of these ancient texts is limited by the inefficiency of information retrieval. Additionally, existing methods for extracting, filtering, and visualizing information from traditional Chinese medicine classics are insufficient to analyze the interconnections within the theoretical knowledge of Chinese medicine. In light of this, we have closely collaborated with domain experts to propose an interactive visual analysis system based on the construction and processing of a knowledge graph for traditional Chinese medicine. This system aids experts in exploration, comprehension, and inference from the data within traditional Chinese medicine classics. It employs an innovative knowledge graph-based layout of medicinal prescriptions, allowing for the exploration of the interrelationships among theories from the perspectives of the holistic view and syndrome differentiation and treatment theory of Chinese medicine. Our system comprises a deep learning-based method for constructing and processing the knowledge graph, an algorithm for discerning the correlation of chief symptoms based on the traditional Chinese medicine theory of 'Jun, Chen, Zuo, and Shi', and a meticulously designed visualization interface. Through case studies, the effectiveness of this system has been verified, receiving positive feedback from domain experts.