Optimal Gamma density to Obfuscate Quantitative data with Added Noise
share
Protecting the privacy of individuals in a data-set is no less important than making statistical inferences from it. In case the data in hand is quantitative, the usual way to protect it is to add a noise to the individual data values. But, what should be an ideal density used to generate the noise, so that we can get the maximum use of the data, without compromising privacy? In this paper, we deal with this problem and propose a method of selecting a density within the Gamma family that is optimal for this purpose.
READ FULL TEXT