Activity Number:
|
239
- Synthetic Data and Differential Privacy: Data, Privacy and the Public Good
|
Type:
|
Invited
|
Date/Time:
|
Tuesday, August 4, 2020 : 1:00 PM to 2:50 PM
|
Sponsor:
|
Survey Research Methods Section
|
Abstract #309230
|
|
Title:
|
Differentially Private Data Synthesis: Challenges and Opportunities
|
Author(s):
|
Fang Liu*
|
Companies:
|
University of Notre Dame
|
Keywords:
|
differential privacy;
data sythesiis;
multiple syntesis;
data privacy;
privacy budget;
utility
|
Abstract:
|
Data synthesis is a statistical disclosure limitation technique that releases synthetic data sets with pseudo individual records so to protect sensitive information of the individuals in the original data . Traditional data synthesis techniques often rely on strong assumptions of a data intruder’s behaviors and background knowledge to assess disclosure risk. Differential privacy formulates a theoretical framework for a strong and robust privacy guarantee in data release without making assumptions about the intruder's behaviors. Efforts have been made aiming to incorporate differential privacy in the data synthesis process. In this talk, I will examine the current practice in DIfferentially Private Data Synthesis (DIPS) techniques, outline the challenges, and list some opportunities for future research to improve the practical feasibility of DIPS and preserve the utility and statistical validity of synthesized data.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|