Releasing Correlated Trajectories: Towards High Utility and Optimal Differential Privacy
Releasing Correlated Trajectories: Towards High Utility and Optimal Differential Privacy
复制标题
释放相关轨迹：迈向高实用性和最佳差异化隐私
DOI：
10.1109/tdsc.2018.2853105
复制
发表时间：
2020-09-01
影响因子：
7.3
通讯作者：
Jia, Xiaohua
中科院分区：
文献类型：
--
作者：
Ou, Lu;Qin, Zheng;Jia, Xiaohua
两个用户的轨迹之间的相互关联对于产品推荐和社交媒体等现实应用非常有用。在提供巨大好处的同时，相关轨迹的公布可能会泄露敏感的社会关系，因为相互关联和社会关系之间存在潜在联系。就我们所知，我们迈出了第一步，提出了一个数学上严格的内联公式&tex-notation=“LaTeX”>；$n$<；/tex-math>；<；alternatives>；<；mml:math>；<；mml:mi>；n<；/mml:mi>；<；/mml:math>；<；inline-graphic xlink:href=“qin-ieq1-2853105.gif”/>；<；/alternatives>；<；/Inline-FORMUM>；-Body Laplace框架，满足<；inline-FORMULE>；<；Tex-Math notation=“LaTeX”>；$\varepsilon$<；/tex-math>；<；alternatives>；<；mml:math>；<；mml:mi>；ɛ<；/mml:mi>；<；/mml:math>；<；inline-graphic xlink:href=“qin-ieq2-2853105.gif”/>；<；/alternatives>；<；/INLINE-FORMUM&>-差异隐私，通过<；INLINE-FORMULE>；<；tex-notation=“LaTeX”>；$n$<；/tex-math>；<；alternatives>；<；mml:math>；<；mml:mi>；n<；/mml:mi>；<；/mml:math>；<；inline-graphic xlink:href=“qin-ieq3-2853105.gif”/>；<；/alternatives>；之间的相互关联有效地阻止社交关系推理<；/inline-公式&>-两个用户的节点轨迹。通过定义轨迹相关性分数来衡量两个用户之间的社会关系，该问题被数学地表达出来。然后，在内联公式<；<；tex-notation=“LaTeX”>；$n$<；/tex-math>；<；alternatives>；<；mml:math>；<；mml:mi>；n<；/mml:mi>；<；/mml:math>；<；inline-graphic xlink:href=“qin-ieq4-2853105.gif”/>；<；/alternatives>；<；/inline-formula>；下-Body Laplace框架下，针对位置距离度量的数据效用和位置相关性度量的数据效用，我们提出了两种基于拉格朗日乘子的差分私有(LMDP)方法来优化隐私预算，即UD-LMDP和UC-LMDP。此外，我们还对隐私、数据效用、对手知识和约束优化进行了详细的分析。最后，我们用真实数据进行了实验研究。实验结果表明，与已有方法相比，本文提出的方法具有更好的保密性和数据利用率。
A mutual correlation between trajectories of two users is very helpful to real-life applications such as product recommendation and social media. While providing tremendous benefits, the releasing of correlated trajectories may leak sensitive social relations, due to potential links between mutual correlations and social relations. To the best of our knowledge, we take the first step to propose a mathematically rigorous <inline-formula><tex-math notation="LaTeX">$n$</tex-math><alternatives><mml:math><mml:mi>n</mml:mi></mml:math><inline-graphic xlink:href="qin-ieq1-2853105.gif"/></alternatives></inline-formula>-body Laplace framework, satisfying <inline-formula><tex-math notation="LaTeX">$\varepsilon$</tex-math><alternatives><mml:math><mml:mi>ɛ</mml:mi></mml:math><inline-graphic xlink:href="qin-ieq2-2853105.gif"/></alternatives></inline-formula>-differential privacy, which efficiently prevents a social relation inference through the mutual correlation between <inline-formula><tex-math notation="LaTeX">$n$</tex-math><alternatives><mml:math><mml:mi>n</mml:mi></mml:math><inline-graphic xlink:href="qin-ieq3-2853105.gif"/></alternatives></inline-formula>-node trajectories of two users. The problem is mathematically formulated by defining a trajectory correlation score to measure the social relation between two users. Then, under the <inline-formula><tex-math notation="LaTeX">$n$</tex-math><alternatives><mml:math><mml:mi>n</mml:mi></mml:math><inline-graphic xlink:href="qin-ieq4-2853105.gif"/></alternatives></inline-formula>-body Laplace framework, we propose two Lagrange Multiplier-based Differentially Private (LMDP) approaches to optimize the privacy budgets, for the data utility measured by location distances and the data utility measured by location correlations, i.e., UD-LMDP and UC-LMDP. Also, we present detailed analyses of privacy, data utility, adversary knowledge and the constrained optimizations. Finally, we perform experimental studies with real-life data. Our experimental results show that our proposed approaches achieve better privacy and data utility than the existing approaches.