Enhancing privacy in participatory sensing applications with multidimensional data

Groat, Michael M.; Edwards, Benjamin; Horey, James; He, Wenbo; Forrest, Stephanie

doi:10.1109/percom.2012.6199861

Cited by 57 publications

(25 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, local privacy from the end user can ensure the consistency of the privacy guarantees when there are multiple accesses to users' data, in contrast to non-local privacy schemes that has to properly split and assign privacy budgets to different steps [5], [21], [35]. In existing work [15][12] [14], local privacy is implemented with randomized response technique [34]. However, the correlations and sparsity in high-dimensional data are not well considered, which will cause low scalability and utility for highdimensional data [25], [35].…”

Section: Privacy In Distributed Settingmentioning

confidence: 99%

See 1 more Smart Citation

High-Dimensional Crowdsourced Data Distribution Estimation with Local Privacy

Ren

et al. 2016

2016 IEEE International Conference on Computer and Information Technology (CIT)

View full text Add to dashboard Cite

Abstract-High-dimensional crowdsourced data collected from numerous users produces rich knowledge for our society. However, it also brings unprecedented privacy threats to the participants. Local privacy, a variant of differential privacy, is proposed to eliminate privacy concerns. Unfortunately, achieving local privacy on high-dimensional crowdsourced data raises great challenges in terms of both computational efficiency and effectiveness. To this end, based on Expectation Maximization (EM) algorithm and Lasso regression, we first propose efficient multi-dimensional joint distribution estimation algorithms that maintain local privacy. Then, we develop a Locally privacy-preserving high-dimensional data Publication algorithm, LoPub, by taking advantage of our distribution estimation techniques. In particular, both correlations and joint distributions among multiple attributes are identified to reduce the dimensionality of crowdsourced data, thus achieving both efficiency and effectiveness in high-dimensional data publication. To the best of our knowledge, this is the first work addressing high-dimensional crowdsourced data publication with local privacy. Extensive experiments on realworld datasets demonstrate that our multivariate distribution estimation scheme significantly outperforms existing estimation schemes in terms of both communication overhead and estimation speed, and confirm that our LoPub scheme can keep average 80% and 60% accuracy over the published approximate datasets in terms of SVM and random forest classification, respectively.

show abstract

Section: Privacy In Distributed Settingmentioning

confidence: 99%

“…However, the participants' privacy can still be easily inferred or identified due to the publication of crowdsourced data [15], [33], especially high-dimensional data, even though some existing privacy-preserving schemes and end-to-end encryption are used. The reasons for privacy leaks are two-fold:…”

Section: Introductionmentioning

confidence: 99%

High-Dimensional Crowdsourced Data Distribution Estimation with Local Privacy

Ren

et al. 2016

2016 IEEE International Conference on Computer and Information Technology (CIT)

View full text Add to dashboard Cite

show abstract

“…Groat et al [10] consider multidimensional data to evaluate the user privacy, i.e., they consider spatio-temporal dimensions, the sensed data and more. But, they do not take into account the continuous data disclosure, which would be disastrous for the users in case of an attack on a multidimensional scale.…”

Section: Related Workmentioning

confidence: 99%

User-side adaptive protection of location privacy in participatory sensing

et al. 2013

View full text Add to dashboard Cite

The participatory sensing paradigm, through the growing availability of cheap sensors in mobile devices, enables applications of great social and business interest, e.g., electrosmog exposure measurement and early earthquake detection. However, users' privacy concerns regarding their activity traces need to be adequately addressed as well. The existing static privacy-enabling approaches, which hide or obfuscate data, offer some protection at the expense of data value. These approaches do not offer privacy guarantees and heterogeneous user privacy requirements cannot be met by them. In this paper, we propose a user-side privacy-protection scheme; it adaptively adjusts its parameters, in order to meet personalized location-privacy protection requirements against adversaries in a measurable manner. As proved by simulation experiments with artificial-and real-data traces, when feasible, our approach not only always satisfies personal location-privacy concerns, but also maximizes data utility (in terms of error, data availability, area coverage), as compared to static privacyprotection schemes.

show abstract

“…These two distributions span a significantly smaller range of utility, 0.786 to 0.802, which corresponds to 1.6% of the privacy metric. Since this effect on utility is so small, Groat et al [46] interpreted utility to be independent of the underlying distribution. This is a reasonable simplification because the number of categories and the number of participants have such a dominant effect on the metric's value.…”

Section: Experimental Study Of Trade-offs Between Privacy and Utilitymentioning

confidence: 99%