Purpose/Significance To construct a knowledge graph for AIDS prevention and control, and to achieve intelligent question and answering(Q&A), so as to provide a scientific basis for AIDS prevention and control, and reduce the disease burden. Method/Process Multi-source heterogeneous information such as expert consensus and diagnosis and treatment guidelines on AIDS prevention and control at home and abroad are systematically sorted out. By leveraging natural language processing (NLP) and big data technologies, combined with prompt word design, entities and the relationships between entities are extracted. Combining knowledge graph (KG), retrieval-augmented generation (RAG) and large language model (LLM), a Q&A system suitable for the field of AIDS prevention and control is constructed. Result/Conclusion The system can enhance the accuracy and effectiveness of Q&A related to AIDS knowledge, providing a feasible path for the intelligent development of AIDS prevention and control.