Implementing Differential Privacy: An evaluation from Developers' Standpoint
Lead Research Organisation:
University of Bristol
Department Name: Computer Science
Abstract
We want to help developers to get Differential Privacy (DP) right.
Balancing privacy and utility has been an area of research when developing Privacy Enhancing Technologies (PETs) to extract valuable information from datasets while preserving privacy controllably. DP is one way to guarantee privacy and mitigate the amount of information disclosed. Existing tools for DP have usability gaps that reduce their adoption by developers. By making DP more usable, we can help drive adoption of the technique.
This PhD aims to:
1. Simplify DP concepts for developers to drive adoption.
2. Understand how DP fits with privacy and regulatory guidelines, and whether it is supported by current implementations.
3. Work out how to fit DP techniques into existing data analysis and machine learning pipelines.
Balancing privacy and utility has been an area of research when developing Privacy Enhancing Technologies (PETs) to extract valuable information from datasets while preserving privacy controllably. DP is one way to guarantee privacy and mitigate the amount of information disclosed. Existing tools for DP have usability gaps that reduce their adoption by developers. By making DP more usable, we can help drive adoption of the technique.
This PhD aims to:
1. Simplify DP concepts for developers to drive adoption.
2. Understand how DP fits with privacy and regulatory guidelines, and whether it is supported by current implementations.
3. Work out how to fit DP techniques into existing data analysis and machine learning pipelines.
Organisations
People |
ORCID iD |