针对垂直划分的分布式数据库提出了一种基于隐私保护的分布式聚类算法PPDC-VP,该算法基于K-Means的思想实现分布式聚类,并且聚类过程中应用扰乱技术保护本站点真实信息不被传送到其它站点,从而达到隐私保护的目的.理论分析和实验结果表明PPDC-VP算法是有效的.
Aiming at the vertically partitioned database, this paper presents a distributed clustering algorithm PPDC _ VP based on privacy-preserving. The algorithm is based on the idea of K-Means to realize distributed clustering, and uses the perturbation technology to protect the real information of the site from being transferred to other sites in clustering procedure. Theoretical analysis and experimental results show that algorithm PPDC_ VP is effective.