An Event Grouping Approach for Infinite Stream with Differential Privacy

Cheng, Mian; Sun, Yipin; Zhao, Baokang; Su, Jinshu

doi:10.1007/978-3-319-49178-3_8

Cited by 5 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, there are subtle differences in the error calculation; in particular, in the selection of the sanity bound of MRE. For instance, [28] uses a data-dependent sanity bound 𝛾, whereas [8] fixes 𝛾 = 1.0. In three publications, the sanity bound is not stated, even though the used streams contain query results of 0, requiring 𝛾 > 0.…”

Section: Error Metrics Set Ementioning

confidence: 99%

“…In the literature, various mechanisms are proposed that sanitize streams to achieve 𝑤-event differential privacy [7,8,21,23,26,28]. All of these mechanisms sanitize the stream by injecting noise into the query results.…”

Section: Introductionmentioning

confidence: 99%

“…Specifically, most studies focus on a small set of varying realworld data streams [7,8,21,23,26,28]. We observe two challenges: First, the selection is hampered by the fact that many streams are not publicly available.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Benchmarking the Utility of w -Event Differential Privacy Mechanisms - When Baselines Become Mighty Competitors

2023

View full text Add to dashboard Cite

The w -event framework is the current standard for ensuring differential privacy on continuously monitored data streams. Following the proposition of w -event differential privacy, various mechanisms to implement the framework are proposed. Their comparability in empirical studies is vital for both practitioners to choose a suitable mechanism, and researchers to identify current limitations and propose novel mechanisms. By conducting a literature survey, we observe that the results of existing studies are hardly comparable and partially intrinsically inconsistent. To this end, we formalize an empirical study of w -event mechanisms by re-occurring elements found in our survey. We introduce requirements on these elements that ensure the comparability of experimental results. Moreover, we propose a benchmark that meets all requirements and establishes a new way to evaluate existing and newly proposed mechanisms. Conducting a large-scale empirical study, we gain valuable new insights into the strengths and weaknesses of existing mechanisms. An unexpected - yet explainable - result is a baseline supremacy, i.e., using one of the two baseline mechanisms is expected to deliver good or even the best utility. Finally, we provide guidelines for practitioners to select suitable mechanisms and improvement options for researchers.

show abstract

Section: Error Metrics Set Ementioning

confidence: 99%