摘要
以准确评估、降低虚拟企业审计风险为目的,设计了基于数据挖掘技术的虚拟企业审计风险评估模型。基于虚拟企业审计风险的主要特征,从重大报错风险、审计人员风险和审计程序风险三方面选取造成虚拟企业审计风险的不同因素作为评估指标,采用层次分析法确定各指标权重值,选取权重值大于阈值的指标构建虚拟企业审计风险评估指标体系。对指标体系内的定性指标与定量指标进行标准化处理,使不同风险评估指标具有可比性。依照评估指标采集相关数据,利用数据挖掘技术中的BP神经网络构建由输入层、中间层和输出层组成的评估模型,基于神经网络学习过程对指标数据进行训练,输出虚拟企业审计风险评估结果。模型测试结果显示,该模型能够准确评估虚拟企业审计风险,基于评估结果改善虚拟企业审计过程薄弱环节,降低虚拟企业审计风险。
In order to accurately assess and reduce the audit risk of virtual enterprise,a virtual enterprise audit risk assessment model based on data mining technology is designed.Based on the main characteristics of audit risk of virtual enterprise,this paper selects different factors that cause audit risk of virtual enterprise as evaluation indexes from three aspects of major error reporting risk,auditor risk and audit procedure risk.AHP is used to determine the weight value of each index,and the index whose weight value is greater than the threshold value is selected to construct the audit risk evaluation index system of virtual enterprise.The qualitative and quantitative indicators in the index system are standardized to make different risk assessment indicators comparable.According to the evaluation index,the system collects the relevant data,uses the BP neural network of data mining technology to build an evaluation model composed of input layer,middle layer and output layer.Based on the neural network learning process,the index data are trained,and the audit risk assessment results of virtual enterprises are output.The model test results show that the model can accurately evaluate the audit risk of virtual enterprise,improve the weak link of audit process,and reduce the audit risk of virtual enterprise based on the evaluation results.
作者
袁阳春
刘森
YUAN Yangchun;LIU Sen(China Southern Power Grid Digital Grid Research Institute Co.Ltd.,Guangzhou 510006,China)
出处
《微型电脑应用》
2021年第9期103-106,共4页
Microcomputer Applications
关键词
数据挖掘
虚拟企业
审计
风险评估
指标体系
神经网络
data mining
virtual enterprise
audit
risk assessment
index system
neural network
