基于大语言模型的审计知识应用研究  被引量:3

Research on the Application of Audit Knowledge Based on Large Language Models

在线阅读下载全文

作  者:万钧 Wan Jun

机构地区:[1]浙江省审计厅计算机审计中心,310007

出  处:《审计研究》2024年第5期38-44,74,共8页Auditing Research

摘  要:随着人工智能技术的不断发展,大语言模型在多个领域逐渐展现出强大的应用潜力。在国家审计领域,审计知识的管理和应用面临知识分散、供需不平衡和敏感信息利用受限等痛点。大语言模型强大的自然语言理解能力和生成能力为审计知识应用提供了新的解决方案,如智能问答、非结构化文档分析等。本文讨论了基于大语言模型的审计知识应用框架设计,通过构建审计知识仓、设计审计知识管理系统和审计知识应用场景,形成一套科学高效的审计知识管理应用体系。通过实践证明,大语言模型技术在审计知识管理应用中,能有效降低人力成本,提高知识的应用绩效,为审计知识应用提供了实践方向。最后从知识仓运维、优化业务应用体系、智慧审计等三个方面提出了完善审计知识管理应用系统探索方向。With the development of artificial intelligence technology,application of large language models show increasing potential in multiple fields.In the field of government audit,the management and application of auditing knowledge face challenges such as knowledge dispersion,imbalance between supply and demand,and restricted utilization of sensitive information.The powerful natural language understanding and generation capabilities of large language models provide new solutions for the application of auditing knowledge,such as intelligent question and answer and unstructured document analysis.This paper discusses the design of an auditing knowledge application framework based on large language models,and forms a set of rational and efficient auditing knowledge management application system by constructing an auditing knowledge warehouse,designing an auditing knowledge management system,and auditing knowledge application scenarios.Through practice,we prove that large language model technology in the application of auditing knowledge management can effectively reduce labor costs,improve the application performance of knowledge,and provide a practical direction for the application of auditing knowledge.Finally,it proposes the exploration direction of improving the auditing knowledge management application system from three aspects:knowledge warehouse operation and maintenance,optimization of business application system,and intelligent auditing.

关 键 词:大语言模型 国家审计 审计知识管理 审计知识应用 知识仓 

分 类 号:TP18[自动化与计算机技术—控制理论与控制工程] F239.4[自动化与计算机技术—控制科学与工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象