期刊文献+

基于差分隐私的企业财务会计数据安全共享方法研究 被引量:3

Research on Enterprise Financial Accounting Data Security Sharing Method Based on Differential Privacy
下载PDF
导出
摘要 针对现有财务会计数据共享中存在开销较大、隐私安全性较低等问题,提出一种基于差分隐私的企业财务会计数据安全共享方法.通过差分隐私技术构建强化学习的形式化模型,在建模过程中引入马尔可夫决策过程,确定会计数据状态,并获得对应最优值函数,完成企业财务会计数据的动态隐私发布;在此基础上,设计会计数据协作服务,该数据协作服务由四种网络实体构成,分别为域密钥生成器、层密钥生成器、根密钥生成器和云服务器,防止企业财务会计数据存储时泄露,实现细粒度访问控制和数据写操作.通过区块链构建企业财务会计数据安全共享模型,模型由以太坊区块链、贡献者节点、访问者节点,以及链上激励模块构成,实现企业财务会计数据安全共享.通过三个实验数据集测试方法性能,测试结果表明:设计方法隐私安全性较高,能够保持较小的数据损失,且设计方法可在较低的开销下实现会计数据安全共享. Aiming at the problems of high cost and low privacy security in the existing financial account⁃ing data sharing,a secure sharing method of enterprise financial accounting data based on differential privacy is proposed.The formal model of reinforcement learning is constructed through differential privacy technology.In the modeling process.Markov decision-making process is introduced to determine the state of accounting data and obtain the corresponding optimal value function to complete the dynamic privacy release of enterprise financial accounting data.On this basis,the accounting data collaboration service is designed.The data collaboration service is composed of four network entities,namely domain key genera⁃tor,layer key generator,root key generator and cloud server,to prevent leakage in enterprise financial ac⁃counting data storage and realize fine-grained access control and data write operation.A secure sharing model of enterprise financial accounting data through blockchain was built.The model is composed of Ethereum blockchain,contributor node,visitor node and incentive module on the chain to realize the secure sharing of enterprise financial accounting data.Through three experimental data sets to test the perfor⁃mance of the method,the test results show that the design method has high privacy security,can maintain small data loss,and the design method can realize the secure sharing of accounting data with low over⁃head.
作者 戴小凤 朱卫东 DAI Xiao-feng;ZHU Wei-dong(Department of Engineering Management,Anhui Audit College,Hefei 230601,China;School of Management,Hefei University of Technology,Hefei 230009,China)
出处 《通化师范学院学报》 2023年第4期88-94,共7页 Journal of Tonghua Normal University
基金 2020年度安徽省教育厅高校人文社会科学研究重点项目(SK2020A0894) 2020年安徽省教育厅质量工程支持疫情防控期间高校线上教学工作特需项目重大线上教学改革研究项目(020zdxsjg130) 2019年度安徽省审计厅重点科研课题(AHSJ-ZB-2019-01)。
关键词 差分隐私 马尔可夫决策过程 企业财务会计数据 数据安全共享 动作值函数 转移概率矩阵 differential privacy markov decision process enterprise financial accounting data data security sharing action value function transition probability matrix
  • 相关文献

参考文献16

二级参考文献146

共引文献200

同被引文献11

引证文献3

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部