|
Abstract:
|
Privacy protection and data security have received a huge amount of attention these days due to the increasing need to protect various sensitive information like credit card data, medical data etc. There are various well-known ways of masking data, such as, Top-coding, Grouping, Adding Noise, Rank Swapping, etc. In Official Statistics, the main goal of most studies is to analyze a data-set to extract different statistics like mean, median, variance etc. which may help in various statistical analyses. But, in case the data is sensitive, it may be completely impossible to publish it in its raw form. In such cases, statistical agencies often release masked version of original data, sacrificing some information. Data Obfuscation refers to the type of data masking where some useful information about the complete data-set remains even after hiding the individual sensitive information. So, the main objectives of Data obfuscation are (i) minimize risk of disclosure resulting from providing access to the data; (ii) maximize the analytic usefulness of the data. The goal is to suggest a procedure with reasonable masking of the data-set which may as well return a good guess of the quantiles.
|
