Construction and Application of an Intelligent Q&A System for Traditional Chinese Medicine Based on Knowledge Graphs and Artificial Intelligence

Abstract

With the deep integration of artificial intelligence (AI) and traditional medicine, the digitization and intelligentization of Traditional Chinese Medicine (TCM) have become important research directions. This paper proposes an intelligent TCM question-answering (QA) system based on knowledge graphs and natural language processing (NLP). By structurally representing TCM knowledge and constructing a multi-dimensional relational graph, and by combining deep learning models, the system enables intelligent QA and personalized health analysis. It integrates classical TCM texts, formulas, medicinal materials, and clinical data, and employs techniques such as BERT fine-tuning and graph neural network (GNN)–based dynamic inference to achieve precise understanding and answering of users’ natural-language questions. Experiments show strong performance in answer accuracy, response speed, and user satisfaction, demonstrating solid practicality and scalability. The study provides a feasible pathway for the modernization and intelligent service of TCM knowledge.