首页> 外文会议>International Conference on Cloud Computing and Security >Conducting Correlated Laplace Mechanism for Differential Privacy
【24h】

Conducting Correlated Laplace Mechanism for Differential Privacy

机译:进行相关的拉普拉斯差异隐私机制

获取原文

摘要

Recently, differential privacy achieves good trade-offs between data publishing and sensitive information hiding. But in data publishing for correlated data, the independent Laplace noise implemented in current differential privacy preserving methods can be detected and sanitized, reducing privacy level. In prior work, we have proposed a correlated Laplace mechanism (CLM) to remedy this problem. But the concrete steps and detailed parameters to imply CLM and the complete proof has not been discussed. In this paper, we provide the complete proof and specific steps to conduct CLM. Also, we have verified the error of our implement method. Experimental results show that our method can retain small error to generate correlated Laplace noise for large quantities of queries.
机译:最近,差异隐私在数据发布和敏感信息隐藏之间实现了良好的折衷。但是在相关数据的数据发布中,可以检测和清除以当前差异隐私保护方法实现的独立拉普拉斯噪声,从而降低了隐私级别。在先前的工作中,我们提出了一种相关的拉普拉斯机制(CLM)来解决此问题。但是还没有讨论暗示CLM的具体步骤和详细参数,以及完整的证明。在本文中,我们提供了进行CLM的完整证明和特定步骤。另外,我们已经验证了实现方法的错误。实验结果表明,对于大量查询,我们的方法可以保留较小的误差以生成相关的拉普拉斯噪声。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号