数据发布中面向多敏感属性的隐私保护技术

Privacy preserving technology for multiple sensitive attributes in data publishing

针对多敏感属性数据发布中存在的隐私泄露问题,在分析多维桶分组技术的基础上,继承了基于有损连接对隐私数据进行保护的思想,提出了一种(g,l)-分组方法,首先对多敏感属性根据各自的敏感度进行分组,然后将分组数作为多维桶的各个维的维数。同时还给出了两种不同的线性时间的分组算法:一般(g,l)-分组算法(GGLG)和最大敏感度优先算法(MSF)。实际数据集上的大量实验结果表明,该方法可以明显地减少隐私泄露,增强数据发布的安全性。

In view of the privacy leak problem of secure data publishing when sensitive data contains multi attributes,based on the multi-dimension bucket grouping approach,this paper proposed a（g,l）-grouping approach on the idea of lossy join.It divided sensitive attributes into groups according to the sensitivity,and set the size of each group as the dimension number of each dimension of the multi-dimension bucket.And proposed two specific line time based（g,l）-grouping algorithms,which were general（g,l）-grouping algorithm（GGLG） and maximal sensitivity first algorithm（MSF）.Experimental results on the real world datasets show that the new model is able to reduce privacy disclosure apparently and enforce security of data publishing.