摘要
在移动轨迹数据发布场景中,由于攻击者所拥有的背景知识及攻击模型的差异,使得无法对不同隐私保护机制的隐私保护度和数据可用性进行统一评价。针对该问题,提出了一个顾及用户、攻击者和隐私保护机制的度量框架。定义了一个融合攻击者背景知识和攻击方法等因素的隐私度量指标Um,使得不同的隐私保护机制在统一度量下能够进行有效性比较,有助于数据发布方选择合适的隐私保护机制以获取隐私保护和数据可用性之间的均衡。在两个真实轨迹数据集上进行的实验验证了该框架以及度量指标的有效性。
The privacy guarantees and data utility provided by various mechanisms of privacy preservation mechanisms are theoretically very different and can’t be directly compared against each other because they are motivated by different adversary models, making varying assumptions about adversary’s background knowledge and intention in mobile trajectory data sharing/publishing. The quantifying framework that includes users, attacks, and privacy preservation mechanism is proposed in this paper. Moreover, a measure that integrated adversary’s background knowledge and models is designed to implement a comparison of the effectiveness of different mechanisms, and help data publishers to choose the appropriate mechanism to obtain a trade-off between privacy and data utility. The effectiveness of the framework and the metric is verified through a set of experiments over two real mobile trajectory datasets.
作者
徐振强
王家耀
崔晓杰
XU Zhenqiang;WANG Jiayao;CUI Xiaojie(Information Engineering University, Zhengzhou 450001, China;College of Information Science and Technology, Henan University of Technology, Zhengzhou 450001, China)
出处
《测绘科学技术学报》
北大核心
2019年第2期196-201,208,共7页
Journal of Geomatics Science and Technology
基金
国家自然科学基金项目(61772173)
河南省高等学校重点科研项目(16A520006
18A520006)
河南省高校科技创新人才支持计划项目(19HASTIT027)
河南工业大学校级基金项目(2017QNJH01)
关键词
轨迹数据发布
隐私保护
隐私度量
数据可用性度量
度量框架
trajectory data publishing
privacy preserving
privacy metric
data utility metric
quantifying framework
