Survey of Differentially Private Accuracy Improving Techniques for Publishing Histograms and Synthetic Data
Poster (konferens), 2019
In this work we focus on the analyses that produce histograms or synthetic data. Histograms and synthetic data are interesting to study because they provide a sanitized form of the original data for which access is restricted. There is growing interest in the scientific community to demonstrate different ways to improve the accuracy of differentially private histograms or synthetic data. However, there are a few work that systematize the knowledge gained by those scientific investigations. Thus our aim is to analyze and structure the state-of-art techniques that improve the accuracy of histograms or synthetic data published under differential privacy. We used the systematic literature review as the research method to summarize the state-of-the-art. Our preliminary result that categorize the state-of-the art is illustrated in Figure 1.
differential privacy
data privacy
systematic literature review (SLR)
systematization of knowledge (SoK)
accuracy improvement
Författare
Jenni Reuben
Karlstads universitet
Boel Nelson
Chalmers, Data- och informationsteknik, Informationssäkerhet
Stockholm, ,
Ämneskategorier
Annan data- och informationsvetenskap
Styrkeområden
Informations- och kommunikationsteknik