A systematic study in intelligent software engineering based on Large Language Model

The application and research of Large Language Model (LLM) in intelligent software engineering have become a hot topic, offering strong theoretical significance and practical value. Researchers in this field have covered various stages of software engineering. However, there is still a lack of comprehensive and holistic research analysis that unifies the current research findings, which hinders the technological development of this field. Addressing this issue, this paper conducts a comprehensive survey of LLM research in the domain of software engineering. We summarize and extract five specific subfields, including LLM and prompts, code analysis, human-computer interaction, fault prediction, and other software engineering aspects. For these representative research directions, it examines significant research works, traces their research context, and discusses their development trends. This work provides a research foundation and contextual support for subsequent software engineering development based on large models.