一种基于差分隐私保护的skyline查询方法

为了解决差分隐私保护机制中重复攻击会泄露用户隐私的问题，提出了一种基于动态页敏感度调节的skyline查询方法. 首先，提出了依据最优主导页的计算页敏感度方法，提高页敏感度计算的效率；其次，为了合理设置隐私预算值，提出了基于置信率的隐私预算值调节方法；最后，基于隐私预算值动态更新查询次数的上界，实现了基于差分隐私保护的skyline查询方法. 实验结果表明:所提出方法在隐私预算值设定小于0.8时，隐私数据的泄露数由787个降低到423个. 

A Skyline Query Method Based on Differential Privacy Protection

In order to solve the problem that replay attacks in the differential privacy protection mechanism will leak user privacy, a skyline query method based on dynamic page sensitivity adjustment is proposed. First, in order to improve the efficiency of page sensitivity calculation, a method for calculating page sensitivity on the basis of the optimal dominant page is presented. Secondly, to reasonably set the privacy budget value, a privacy budget value adjustment method based on the confidence rate is developed. Finally, the upper bound of query times is dynamically updated based on the privacy budget value, and the skyline query method based on differential privacy protection is realized. The experimental results show that the proposed method reduces the number of leaked private data from 787 to 423 when the privacy budget value is set to be less than 0.8.