Formal Privacy Guarantees and Analytical Validity of OnTheMap Public-Use Data

John Abowd
Cornell University

Formal privacy models and statistical disclosure limitation address similar issues. Application of formal privacy models to general purpose official government statistical releases presents challenges in preserving the usefulness of the released data. This paper considers a detailed example of the application of probabilistic differential privacy to a real Census Bureau product (OnTheMap). We discuss the statistical analysis and decision-making process for choosing the privacy parameters and assessing analytical validity.

Presentation (PDF File)

Back to Statistical and Learning-Theoretic Challenges in Data Privacy