This lecture examines, from a technological angle, the problems involved with gathering, storing and accessing data about individuals in databases whilst preserving their individual privacy. Starting with basic computer-science definitions of privacy, this lecture explores early work in reidentifying individuals and their sensitive characteristics in apparently anonymous and anonymised data. It presents early approaches used to prevent such reidentification, the limitations of these methods, and examples of high-profile reidentifications performed on real-world data. Finally the current state of the art, differential privacy, is introduced and explained, along with its strengths, limitations and practical applications.”
