基于垂直划分的隐私保护skyline查询

Privacy-preserving Skyline Processing on Vertically-Partitioned Data

Skyline查询是一种重要的数据分析方法,在推荐系统中有着广泛的应用。近年来,随着隐私保护需求的不断增长,分布式数据集上的隐私保护skyline查询问题受到越来越多的关注。然而,现有的分布式数据集上的隐私保护skyline查询方案大多只适用于水平分布数据集,不能满足垂直分布数据集上的skyline查询需求。为此,深入研究了垂直分布式数据集上保护隐私的skyline查询问题,提出了一种基于保序加密的垂直分布数据集上的隐私保护skyline查询算法,可以在保护数据隐私的同时,有效支持skyline查询过程。理论分析证明了提出协议的正确性和安全性,并通过理论分析和模拟实验对协议运行效率进行了评估,结果显示新方案具有较高的运行效率。

As a method of data analysis,skyline query plays an important role in many real-world applications,such as rec- ommender system.Recently,with the growth of privacy concerns,many schemes have been proposed to achieve privacy-preserving skyline query on distributed databases.Nevertheless,most of them focus on horizontally-partitioned dataset,and cannot support se- cure skyline query on vertically-distributed databases.In this paper,we focus on privacy-preserving skyline query on vertically- partitioned data and propose an efficient scheme based on order-preserving encryption tor it.In the proposed scheme,we can guar- antee the privacy of each data.We theoretically prove the security of our scheme.Additionally,we leverage extensive experiments to evaluate our proposed method, which shows our scheme can achieve high efficiency.