Research on Construction and Application of Knowledge Graph in Science and Technology Field Based on Large Language Model

In recent years, large language models have been widely used in the field of knowledge graph. The construction of traditional knowledge graph depends on the experience of domain experts, which is time-consuming and inefficient. Therefore, This paper proposes a construction method that combines intelligent agents with thought chains and self verification, by gradually guiding large models to extract knowledge and verifying it with other models. On this basis, QLoRA is used to fine tune the model. Realize an efficient and automated knowledge graph building process. Invite domain experts to verify the constructed map, so as to ensure the accuracy and integrity of the knowledge graph in semantics and structure.