Implicit privacy preservation: a framework based on data generation
Implicit privacy preservation: a framework based on data generation
复制标题
DOI：
10.1051/sands/2022008
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Qing Yang;Cheng Wang;Teng Hu;Xue Chen;Changjun Jiang
中科院分区：
文献类型：
--
作者：
Qing Yang;Cheng Wang;Teng Hu;Xue Chen;Changjun Jiang
本文讨论了一种特殊的和不可感知的隐私，称为隐式隐私。与传统的（显式）隐私相比，隐式隐私具有两个基本属性：（1）它最初不被定义为隐私属性;（2）它与隐私属性密切相关。也就是说，攻击者可以利用它以一定的概率推断隐私属性，间接导致隐私信息的泄露。针对隐式隐私泄露问题，给出了一个可度量的隐式隐私定义，并提出了一个基于数据生成的事前隐式隐私保护框架IMPOSTER。该框架由隐式隐私检测模块和隐式隐私保护模块组成。前者使用归一化互信息来检测与传统隐私属性密切相关的隐式隐私属性。基于数据生成的思想，后者为生成对抗网络（GAN）框架配备了额外的过滤器，用于消除传统隐私属性与隐式属性之间的关联。我们阐述了理论分析的收敛性的框架。实验结果表明，使用学习生成器，IMPOSTER可以减少隐式隐私的泄露，同时保持良好的数据效用。
This paper addresses a special and imperceptible class of privacy, called implicit privacy. In contrast to traditional (explicit) privacy, implicit privacy has two essential properties: (1) It is not initially defined as a privacy attribute; (2) it is strongly associated with privacy attributes. In other words, attackers could utilize it to infer privacy attributes with a certain probability, indirectly resulting in the disclosure of private information. To deal with the implicit privacy disclosure problem, we give a measurable definition of implicit privacy, and propose an ex-ante implicit privacy-preserving framework based on data generation, called IMPOSTER. The framework consists of an implicit privacy detection module and an implicit privacy protection module. The former uses normalized mutual information to detect implicit privacy attributes that are strongly related to traditional privacy attributes. Based on the idea of data generation, the latter equips the Generative Adversarial Network (GAN) framework with an additional discriminator, which is used to eliminate the association between traditional privacy attributes and implicit ones. We elaborate a theoretical analysis for the convergence of the framework. Experiments demonstrate that with the learned generator, IMPOSTER can alleviate the disclosure of implicit privacy while maintaining good data utility.