About the project
Project Name: DPella
Data analyses with privacy in mind
Team: Alejandro Russo, Marco Gaboardi, Carola Compá
Large amounts of data are being collected about individuals by a variety of systems. Much of the collected data is private: it contains details about individuals and their behaviour. Privacy concerns about individuals restrict the way this information can be exploited, monetized, or released, where utilizing individuals’ data is therefore curbed for ethical, legal, or business reasons.
Ensuring data privacy is a hard problem. It cannot be achieved with a few hacks or as an afterthought. We believe that data privacy needs to be approached in a principle and scientific manner by using mathematics to provide strong guarantees – and this is the philosophy that we follow as computer scientists who are experts in the topic.
In this project, we will work with our privacy-enhancing technology DPella, a tool that perturbs data analyses’ results with carefully calibrated randomized noise to protect the privacy of individuals with mathematical guarantees, while providing information about the accuracy of the results. This project explores the possibilities for our technology to enable governments and companies to produce open, secure, and public data analyses from private datasets. We expect to start liberating the insights provided by private data; insights that otherwise would remain close due to privacy restrictions.