An optimization-based mechanism 🧬 🔐 to selectively hide the minimum number of overlapping SNPs among the family members 👨‍👩‍👧‍👦 who participated in the genomic studies (i.e. GWAS). Our goal is to distort the dependencies among the family members in the original database for achieving better privacy without significantly degrading the data utility.

Master Thesis on Generating Synthetic Data and Disclosure Control of Global COVID-19 Trends and Impact Survey Microdata and Opendata, especially focusing on the evaluation of different synthetic datasets.

This repository contains Python scripts to identify attributes in a dataset and subsequently determine the best QID dimension based on privacy gain and non-uniform entropy.