Research on Optimizing Classification Standards for Immediate Response to Public Complaints Driven by Generative Artificial Intelligence

Beijing Municipal Social Science Fund Planning Project (General Category) (2025-2027), In Progress

This study examines classification standards for digital government service tickets. By integrating top-down design with bottom-up real-world ticket data and leveraging intelligent technology, it constructs an integrated paradigm capable of continuously iterating to optimize the scientific rigor, dynamic adaptability, and processing efficiency of government service ticket classification. This approach is expected to enhance the precision of urban governance and the effectiveness of government services. Academic Value: Distinguishing itself from previous social science fund projects on “immediate response to public complaints” that focused on governance models and process efficiency optimization, this research centers on the shortcomings of existing government service ticket classification standards. The proposed integrated paradigm aims to overcome limitations of traditional classification methods, enabling more precise understanding of public demands, enhancing government service efficiency, and improving decision-making rationality. It serves as a valuable contribution to constructing a refined urban governance paradigm with Chinese characteristics. Applied Value: This research addresses pressing concerns in “immediate response to complaints” reforms and megacity governance, aiming to resolve practical issues such as imprecise classification, inaccurate attribution, and disconnect between responsibilities and accountability. It offers valuable insights for optimizing work order systems nationwide.

This study aims to enhance the scientific rigor, dynamic adaptability, and processing efficiency of government service ticket classification. It begins with a systematic diagnosis of existing classification standards, revealing their inherent limitations and practical bottlenecks. It then introduces large language model (LLM) technology to construct a tagging system capable of dynamically capturing deep semantic meanings. Through expert validation and feedback, the integrated standards and tags form a new classification paradigm for government services, which undergoes iterative refinement. This approach explores a systematic methodology and application framework for leveraging LLMs in optimizing digital government service ticket classification standards.