Releasing data containing sensitive information has an implicit risk that confidential information about individuals became revealed. Perturbative masking methods propose the distortion of the original data sets before publication, in order to obtain a tradeoff between data utility (low information loss) and protection against disclosure (low disclosure risk). In this paper, we empirically evaluate a particular collection of perturbative masking methods for continuous microdata based on the image compression standards JPEG and JPEG 2000.
展开▼