面向隐私保护聚类的平面反射数据扰动方法

Plane reflection data perturbation method for privacy preserving clustering

为了解决聚类挖掘中的隐私保护问题,针对现有的几何数据转换方法隐私保护度低的不足,提出了一种基于平面反射的几何数据转换方法,即任意选择平面上的一条直线,且将所有属性两两配对以构成平面上的点,对每个点作关于直线的对称点,所得数据即转换后的数据。通过实验证明,这种方法简单易行且比平移、缩放、旋转等几何数据转换方法具有更高的隐私保护度。

In order to solve the problem of privacy protection in cluster mining, in view of the low privacy of the existing methods of geometric data transformation, this paper presents a method of plane reflection based geometric data transformation. A straight line in the plane is chosen arbitrarily and all attributes are paired to form points on the plane. The symmetry point of each point is computed. The computed result is the transformed data. Proved by the experiments, this method is simple and has a higher degree of privacy protection than translation, scaling, rotation and other geometric data transformation methods.