
金融与风险管理
Finance and Risk Management
- Publisher:Eastern Sunshine Publishing House Limited
- ISSN Print:3080-9320(P)
- ISSN Online:3080-9339(O)
- Subject Category:Economics & Management
- Publication Frequency:Monthly
- Submissions:1
- Views:273
Related
- Coupling Coordination and Governance Risks of Public Sports Services in the Chengdu-Chongqing Economic Circle
- Research on the Innovative Teaching Path of Exhibition Marketing Course in Higher Vocational Colleges — A Case Study of Nanchong Vocational College of Culture and Tourism
- Comparison of Carbon Disclosure Quality among Leading Petrochemical Enterprises under the TCFD Framework: A Case Study of Sinopec and Hengli Petrochemical
- Research and Exploration on the Exchange and Learning Mechanism for Young Cadres in State-Owned Enterprises
- Research on Wellness Talent Training Path of Higher Vocational Hotel Majors Under the Upgrade of Wellness Cultural Tourism in the Yangtze River Delta
大数据审计中算法线索向审计证据转化的机制与质量控制
Mechanisms and Quality Control for Converting Algorithm-Generated Leads into Audit Evidence in Big Data Auditing
Published: 2026-11-17
Author: 陈思源 :成都市金牛区审计局 四川成都; CHEN Siyuan :Jinniu District Audit Bureau, Chengdu;
Chinese Abstract: 算法线索能否有效转化为审计证据,是影响大数据审计质量的重要问题。本文采用文献分析与规范分析方法,以国家审计为研究对象,构建数据溯源、规则论证、交叉验证与结论约束相衔接的分析框架。研究认为,算法线索的证明力取决于数据适用性、计算条件与审计标准的对应关系及核查程序的有效性。应根据线索性质实行差异化取证,将技术与业务复核、未命中记录检查及异议处理纳入审计流程,以证据质量和核查成本共同评价技术应用效果。
Abstract: The effective conversion of algorithm-generated leads into audit evidence is a critical issue affecting the quality of big data auditing. Using literature review and normative analysis, this study focuses on government auditing and develops an analytical framework integrating data provenance, rule justification, cross-verification, and constraints on audit conclusions. The study argues that the evidential value of algorithm-generated leads depends on data fitness for purpose, the alignment between computational criteria and audit standards, and the effectiveness of verification procedures. Evidence-gathering procedures should be tailored to the nature of each lead. Technical and audit reviews, checks of unflagged records, and the handling of objections should be incorporated into the audit process. The effectiveness of technological applications should be evaluated in terms of both evidence quality and verification costs.
Chinese Keywords: 大数据审计;算法线索;电子审计证据;职业判断;质量控制
Keywords: big data auditing; algorithm-generated leads; electronic audit evidence; professional judgment; quality control
