Dynamic Data Histogram Publishing Based on Differential Privacy
Citations Over TimeTop 22% of 2018 papers
Abstract
Differential privacy, due to its rigorous mathematical proof and strong privacy guarantee, has become a standard for the release of statistics on privacy protection. In the process of its continuous development, many data publishing algorithms that satisfy the differential privacy histogram are proposed. However, most of these algorithms are focused on the release of static data and less research on dynamic data release. A direct way of publishing dynamic data is to publish a histogram that satisfies the differential privacy at every time point, but this method can lead to high cumulative error and reduce the utility of datasets. In order to solve these problems, we propose a histogram publishing algorithm for differential privacy dynamic data based on Kullback-Leibler(KL) divergence. The algorithm uses KL divergence to calculate the difference between two adjacent data updates. At the same time, for the different values calculated by KL divergence, we adopt three strategies for dynamic data publishing. Extensive experiments on real datasets demonstrate that our algorithm can reduce noise errors and achieves better utility than existing state-of-the-art algorithms.
Related Papers
- → Privacy-preserving governmental data publishing: A fog-computing-based differential privacy approach(2018)48 cited
- → Differentially private data publishing for arbitrarily partitioned data(2020)12 cited
- → Adaptive Differential Privacy Interactive Publishing Model Based on Dynamic Feedback(2018)4 cited
- → Differentially Private Query Learning: from Data Publishing to Model Publishing(2017)
- → A Unifying Privacy Analysis Framework for Unknown Domain Algorithms in Differential Privacy(2023)