Skip to main navigation Skip to search Skip to main content

From reactive to proactive policy implement method through intelligent enterprise matching

  • Yongkang Duan
  • , Guangyu Zhao
  • , Qian Geng
  • , Ping Ji
  • , Jian Jin
    • Beijing Normal University
    • Beijing University of Posts and Telecommunications
    • Department of Management Science and Engineering

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Despite the availability of pro-business policies, many micro, small and medium-sized enterprises (MSMEs) struggle to effectively utilise these resources due to limited capacity. This study proposes an policy-enterprise matching framework that transforms ‘enterprises seeking policies’ approach into a ‘policies seeking enterprises’ mechanism. The mechanism uses large language models (LLMs) as tool for extracting policy restrictive clauses and generating compliance checking strategies. Meanwhile, machine learning methods are used for enterprise risk prediction. The matching framework broadens the research perspective of policy analysis and risk prediction, and provides government staff with decision support based on risk prediction.

    Original languageBritish English
    Article number2628701
    JournalEnterprise Information Systems
    Volume20
    Issue number3
    DOIs
    StatePublished - 2026

    Keywords

    • cross-region recommendation
    • enterprise risk prediction
    • large language models
    • micro, small and medium-sized enterprises
    • Policy-enterprise matching
    • retrieval-augmented generation

    Fingerprint

    Dive into the research topics of 'From reactive to proactive policy implement method through intelligent enterprise matching'. Together they form a unique fingerprint.

    Cite this