Differentially Private Data Release through Multidimensional Partitioning

Xiao, Yonghui; Xiong, Li; Yuan, Chun

doi:10.1007/978-3-642-15546-8_11

Cited by 157 publications

(175 citation statements)

References 33 publications

Supporting

Mentioning

175

Contrasting

Order By: Relevance

“…The usual approach to releasing differentially private data sets is based on histogram queries [31,32], that is, on approximating the data distribution by partitioning the data domain and counting the number of records in each partition set. To prevent the counts from leaking too much information they are computed in a differentially private manner.…”

Section: Related Work On Differentially Private Data Publishingmentioning

confidence: 99%

“…Moreover, since many works on differential privacy focus on preserving the utility of counting queries [35,31,32,27,28,29,30], we measured how the methods preserve the data distribution by building histograms of each attribute and comparing the distribution between the original and masked values according to the well-known Jensen-Shannon divergence (JSD) [56], which is symmetric and bounded in the 0..1 range. At a data set level, we averaged the divergence of all the attributes.…”

Section: Evaluation Measures and Experimentsmentioning

confidence: 99%

See 1 more Smart Citation

Utility-preserving differentially private data releases via individual ranking microaggregation

2016

View full text Add to dashboard Cite

Being able to release and exploit open data gathered in information systems is crucial for researchers, enterprises and the overall society. Yet, these data must be anonymized before release to protect the privacy of the subjects to whom the records relate. Differential privacy is a privacy model for anonymization that offers more robust privacy guarantees than previous models, such as k-anonymity and its extensions. However, it is often disregarded that the utility of differentially private outputs is quite limited, either because of the amount of noise that needs to be added to obtain them or because utility is only preserved for a restricted type and/or a limited number of queries. On the contrary, k-anonymity-like data releases make no assumptions on the uses of the protected data and, thus, do not restrict the number and type of doable analyses. Recently, some authors have proposed mechanisms to offer general-purpose differentially private data releases. This paper extends such works with a specific focus on the preservation of the utility of the protected data. Our proposal builds on microaggregation-based anonymization, which is more flexible and utility-preserving than alternative anonymization methods used in the literature, in order to reduce the amount of noise needed to satisfy differential privacy. In this way, we improve the utility of differentially private data releases. Moreover, the noise reduction we achieve does not depend on the size of the data set, but just on the number of attributes to be protected, which is a more desirable behavior for large data sets. The utility benefits brought by our proposal are empirically evaluated and compared with related works for several data sets and metrics.

show abstract

Section: Related Work On Differentially Private Data Publishingmentioning

confidence: 99%

Section: Evaluation Measures and Experimentsmentioning

confidence: 99%

Utility-preserving differentially private data releases via individual ranking microaggregation

2016

View full text Add to dashboard Cite

show abstract

“…We propose to demonstrate DObjects+, a scalable and extensible framework that is aimed to enable privacy preserving data federation services. The framework extends our DObjects architecture [6], [8] with our ongoing work on distributed anonymization protocols [7], [5], [18] and secure query processing protocols [9] for a seamless access to distributed and possibly private data. We summarize the contributions of the demonstrated framework below.…”

Section: Contributionsmentioning

confidence: 99%

“…While the framework is orthogonal to different privacy principles, we studied several representative state-ofthe-art privacy principles within our framework including ldiversity [14], t-closeness [12], and differential privacy [3], [10]. We show the implications of adopting them in the distributed setting with respect to the above attack space and integrated new or modified notions and algorithms in our framework [7], [5], [18].…”

Section: Contributionsmentioning

confidence: 99%

“…This notion is extremely advantageous in the distributed setting where multiple data providers can use their own data as background knowledge and collude with each other, which has to be explicitly modeled when using adversarial privacy principles. To support non-interactive data release with differential privacy, our framework currently uses an adaptive multi-dimensional partitioning strategy for releasing histograms and an optional set of synthetic data generated from the histograms [18].…”

Section: B Distributed Anonymizationmentioning

confidence: 99%

See 1 more Smart Citation

DObjects+: Enabling Privacy-Preserving Data Federation Services

Jurczyk

Xiong

Goryczka

2012

2012 IEEE 28th International Conference on Data Engineering

Self Cite

View full text Add to dashboard Cite

Abstract-The emergence of cloud computing implies and facilitates managing large collections of highly distributed, autonomous, and possibly private databases. While there is an increasing need for services that allow integration and sharing of various data repositories, it remains a challenge to ensure the privacy, interoperability, and scalability for such services. In this paper we demonstrate a scalable and extensible framework that is aimed to enable privacy preserving data federations. The framework is built on top of a distributed mediatorwrapper architecture where nodes can form collaborative groups for secure anonymization and secure query processing when private data need to be accessed. New anonymization models and protocols will be demonstrated that counter potential attacks in the distributed setting.

show abstract

Differential privacy location data release based on quadtree in mobile edge computing

Liu

Tang

Wan

et al. 2020

Trans Emerging Tel Tech

View full text Add to dashboard Cite

To preserve user privacy in the release of location data, some researchers have proposed a distributed data release framework and a location data release algorithm based on mobile edge computing. This framework can preserve location privacy even if location-based service providers are not trusted. However, the proposed algorithm fails to balance the noise error and the uniform hypothesis error and fails to take the query consistency constraint into account, so query accuracy needs to be improved. Therefore, we follow this framework and propose a differential privacy quadtree partitioning algorithm based on data uniformity heuristic adjustment to further improve query accuracy. First, a differential privacy complete quadtree is constructed, and then the quadtree is heuristically adjusted based on the predefined noise count threshold and the data uniformity threshold to balance the two types of error. Finally, query consistency processing is performed to further improve query accuracy. Experiments based on real-world datasets are conducted to study the impact of two thresholds on query accuracy. Compared with the previous quadtree-based differential privacy partitioning algorithm, our proposed algorithm has higher query accuracy while preserving location privacy. In addition, to address the problem of location privacy leakage in the query process, a range counting query framework based on the Hilbert curve is proposed to preserve location privacy in the query process.

show abstract

Differentially Private Data Release through Multidimensional Partitioning

Cited by 157 publications

References 33 publications

Utility-preserving differentially private data releases via individual ranking microaggregation

Utility-preserving differentially private data releases via individual ranking microaggregation

DObjects+: Enabling Privacy-Preserving Data Federation Services

Differential privacy location data release based on quadtree in mobile edge computing

Contact Info

Product

Resources

About