A Study of the Characteristics of a Differential Privacy Implementation

Niveah T Abraham, Purdue University


Building on the theoretic strength of differential privacy, researchers at the University of California Berkeley have built a tool called FLEX that incorporates differential privacy and have tested FLEX using a dataset based on actual queries submitted by Uber’s employees. In their research, a novel idea called elastic sensitivity was introduced to implement differential privacy; this research in turn tests these concepts on datasets from several domains including automobile, medical, network, password, and social media areas. This research also checks for statistical information gathering accuracy from the datasets and observes performance overhead for queries. The study also presents the results obtained—before and after corrections to the FLEX code—during the analysis of the different datasets to obtain differentially private results.




Springer, Purdue University.

Subject Area

Information Technology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server