Dimensionality Reduction for Privacy Preserving Data Mining using Random Projection Perturbation Approach in Outsourced Environment

Abstract: Data perturbation is a popular technique in privacy-preserving data mining. A major challenge in data perturbation is to balance privacy protection and data utility, which are normally considered as a pair of conflicting factors. We discuss that selectively preserving the task/model specific information in perturbation will help achieve better privacy guarantee and better data utility. One type of such information is the multidimensional geometric information, which is implicitly utilized by many data mining models. To preserve this information in data perturbation, we propose the Random Projection Data Perturbation (RPP) method. In this paper, we describe several aspects of the RPP method. Methods of dimensionality reduction provide a way to understand and visualize the structure of complex data. Recently, they have been proposed for ensuring that a given data, in a lower space, are protected against privacy threats, and meanwhile expose many of the useful and interesting properties of the original data. Dimensionality reduction methods assume that the data records are represented as vectors in a multidimensional space where each dimension represents a single attribute.

Keywords: Privacy-preserving Data Mining, Data Perturbation, Random Projection Perturbation, Privacy Evaluation, Data Mining Algorithms