面向敏感值的层次化多源数据融合隐私保护

数据融合技术能够使用户得到更全面的数据以提供更有效的服务。然而现有的多源数据融合隐私保护模型没有考虑数据提供者的重要程度,以及数据不同属性和属性值的敏感度。针对上述问题,提出了一种面向敏感值层次化的隐私模型,该模型通过数据提供者对数据的匿名程度要求来设置数据属性以及属性值的敏感度以实现敏感值的个性化隐私保护。同时结合k-匿名隐私模型以及自顶向下特殊化TDS的思想提出了一种面向敏感值的多源数据融合隐私保护算法。实验表明,该算法既能实现数据的安全融合,又能获得更好的隐私保护。

Hierarchical Privacy Protection of Multi-source Data Fusion for Sensitive Value

Data fusion technology enables users to get more comprehensive data to provide more effective service.Howe-ver,the existing multi-source data fusion privacy protection models do not consider the importance of the data provi-ders,and the sensitivity of different attributes and attribute values.According to the above problems,this paper proposed a hierarchical privacy model for sensitive value.The model enables data providers to set sensitive value of data attributes and attribute values by anonymous degree requirements to realize the individual privacy protection of sensitive values.At the same time,this paper proposed a multi-source data fusion privacy protection algorithm for sensitive value combining with k-anonymous privacy model and the top-to-down specialization TDS.Experiments show that the proposed algorithm can not only realize data security fusion,but also obtain better privacy protection.