摘要
面对5G、人工智能、区域链等技术快速发展的时代,智能科技开始进入大众视野,通过聚类算法对自动化审计进行探究,聚类分析审计的方法能够帮助企业对于盈利表、财产权益变动表、资产负债表、现金流表格等进行分类,同时系统能够自行算出结果。在数据计算的便捷性与准确性上来说都具有强大的优势,针对聚类分析审计体系中可能会有的误差进行分析,最终发现得到的数据结果非常准确,在计算资产的使用率以及市场流动率中会有略微误差,但是整体的误差率不高于审计体系5%的目标,由此可以得知聚类分析审计方法计算的结果较为精准。以商业银行审计项目为案例,评测聚类算法效果。结果表明:使用K-means聚类算法提高审计分析质量可行。
Based on the existing research, facing the intelligent+era composed of the rapid development of 5g, Internet of things, artificial intelligence and blockchain technology,this paper explores the research on automatic audit based on clustering algorithm. The cluster analysis audit system sets classification queries for the accounts in the audited enterprise’s profit statement, balance sheet, statement of changes in owner’s equity and cash flow statement, and the system will automatically generate query results, which are accurate Authenticity is more accurate and efficient than traditional data processing methods. By analyzing the data analysis error level of the audit system based on cluster analysis, the measurement error result of equity multiplier is more stable, the test error of P/E ratio and total asset turnover rate fluctuates to a certain extent, but the test error rate is lower than the test target of 0.5% of the audit system, and the cluster analysis algorithm budget statement audit system has a good effect. Taking the audit project of commercial bank as an example, the effect of clustering algorithm is evaluated. The results show that it is feasible to use k-means clustering algorithm to improve the quality of audit analysis.
作者
边莹
张世国
BIAN Ying;ZHANG Shi-guo(School of accounting,Dalian University of Finance and economics,Dalian 116622,Liaoning,China;Liaoning Zerunxin Certified Public Accountants,Dalian 116000,Liaoning,China)
出处
《贵阳学院学报（自然科学版）》
2022年第2期15-19,共5页
Journal of Guiyang University：Natural Sciences
基金
辽宁省教育厅科学研究经费项目“智能+时代审计自动化发展路径研究”(项目编号:SYDL202009)
大连财经学院青年学术骨干培育计划资助。
关键词
聚类算法
审计
智能
测试误差
自动化
Clustering algorithm
Audit
Intelligence
Test error
Automation
