Study Overview
On May 15, 2018, the European Union adopted the General Data Protection Regulation (GDPR) to protect the privacy and security of EU citizens’ data. This regulation lays out the requirements that data holders -- such as researchers, humanitarian organizations, and financial institutions -- must abide by when collecting, storing, and sharing data. In particular, the GDPR specifies three types of privacy risks that must be ameliorated prior to the sharing of personal information: singling-out attacks, linkage attacks, and attribute inference attacks. While differential privacy -- a state-of-the-art approach for privacy preserving data analysis -- can provably prevent all three attacks (when the privacy parameters are small), less is known about how other privacy-enhancing technologies fare against attribute inference attacks.
Study Results
This project will develop new methods to protect personal data from attribute inference attacks, while also facilitating key downstream use cases.
IBSI Funding Acknowledgement: Lab for Inclusive FinTech (LIFT)