Heavy Hitter Estimation over Set-Valued Data with Local Differential Privacy

Qin, Zhan; Yang, Yin; Yu, Ting; Khalil, Issa; Xiao, Xiaokui; Ren, Kui

doi:10.1145/2976749.2978409

Cited by 276 publications

(197 citation statements)

References 25 publications

Supporting

Mentioning

193

Contrasting

Unclassified

Order By: Relevance

“…the Johnson-Lindenstrauss lemma) to build a primitive to estimate the weight of a single point in the distribution; this is used to find all heavy hitters. Qin et al adapt this approach to the related problem of identifying heavy hitters within set valued data [36]; Chen et al use it on spatial data to build user movement models under LDP [8]. Wang et al describe optimizations that consider asymmetric randomized response and hashing to reduce variance [41].…”

Section: Local Differential Privacy (Ldp)mentioning

confidence: 99%

Marginal Release Under Local Differential Privacy

Cormode

Kulkarni

Srivastava

2018

Proceedings of the 2018 International Conference on Management of Data

128

102

View full text Add to dashboard Cite

Many analysis and machine learning tasks require the availability of marginal statistics on multidimensional datasets while providing strong privacy guarantees for the data subjects. Applications for these statistics range from finding correlations in the data to fitting sophisticated prediction models. In this paper, we provide a set of algorithms for materializing marginal statistics under the strong model of local differential privacy. We prove the first tight theoretical bounds on the accuracy of marginals compiled under each approach, perform empirical evaluation to confirm these bounds, and evaluate them for tasks such as modeling and correlation testing. Our results show that releasing information based on (local) Fourier transformations of the input is preferable to alternatives based directly on (local) marginals.

show abstract

Section: Local Differential Privacy (Ldp)mentioning

confidence: 99%

Marginal Release Under Local Differential Privacy

Cormode

Kulkarni

Srivastava

2018

Proceedings of the 2018 International Conference on Management of Data

128

102

View full text Add to dashboard Cite

show abstract

“…Bassily and Smith theoretically proposed a protocol for frequency estimation and identifying heavy hitters in local DP model, by using the succinct histogram (SH) to represent the data. Considering the heavy hitters over the set‐valued data, the LDPMiner is based on sampling RAPPOR and sampling SH, and consists of two phases to optimize the estimation procedure. Under the data collection scenario of single binary attribute and multiple polychotomous attributes, Wang et al studied the relationship between Laplace mechanism and RR by comparing their theoretical utility error and showing the ε ‐differential privacy satisfaction results of RR.…”

Section: Preliminaries and Related Workmentioning

confidence: 99%

Locally private Jaccard similarity estimation

Yan

Ren

et al. 2018

Concurrency and Computation

View full text Add to dashboard Cite

Summary Jaccard Similarity has been widely used to measure the distance between two sets (or preference profiles) owned by two different users. Yet, in the private data collection scenario, it requires the untrusted curator could only estimate an approximately accurate Jaccard similarity of the involved users but without being allowed to access their preference profiles. This paper aims to address the above requirements by considering the local differential privacy model. To achieve this, we initially focused on a particular hash technique, MinHash, which was originally invented to estimate the Jaccard similarity efficiently. We designed the PrivMin algorithm to achieve the perturbation of MinHash signature by adopting Exponential mechanism and build the Locally Differentially Private Jaccard Similarity Estimation (LDP‐JSE) protocol for allowing the untrusted curator to approximately estimate Jaccard similarity. Theoretical and empirical results demonstrate that the proposed protocol can retain a highly acceptable utility of the estimated similarity as well as preserving privacy.

show abstract

“…Although differential privacy has received considerable interest in the literature [25]- [35], it has been observed by Kifer and Machanavajjhala [11] (see also [12]- [18]) that differential privacy may not work as expected when the data tuples have dependencies. To extend differential privacy for correlated data, prior studies have investigated various privacy metrics [13]- [18], [36].…”

Section: Related Workmentioning

confidence: 99%

Composition properties of Bayesian differential privacy

Zhao¹

2017

2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC)

View full text Add to dashboard Cite

Differential privacy is a rigorous privacy standard that has been applied to a range of data analysis tasks. To broaden the application scenarios of differential privacy when data records have dependencies, the notion of Bayesian differential privacy has been recently proposed. However, it is unknown whether Bayesian differential privacy preserves three nice properties of differential privacy: sequential composability, parallel composability, and post-processing. In this paper, we provide an affirmative answer to this question; i.e., Bayesian differential privacy still have these properties. The idea behind sequential composability is that if we have m algorithms Y1, Y2, . . . , Ym, where Y is independently -Bayesian differential private for = 1, 2, . . . , m, then by feeding the result of Y1 into Y2, the result of Y2 into Y3, and so on, we will finally have an m =1 -Bayesian differential private algorithm. For parallel composability, we consider the situation where a database is partitioned into m disjoint subsets. The -th subset is input to a Bayesian differential private algorithm Y , for = 1, 2, . . . , m. Then the parallel composition of Y1, Y2, . . ., Ym will be max m =1 -Bayesian differential private. The postprocessing property means that a data analyst, without additional knowledge about the private database, cannot compute a function of the output of a Bayesian differential private algorithm and reduce its privacy guarantee.

show abstract

Heavy Hitter Estimation over Set-Valued Data with Local Differential Privacy

Cited by 276 publications

References 25 publications

Marginal Release Under Local Differential Privacy

Marginal Release Under Local Differential Privacy

Locally private Jaccard similarity estimation

Composition properties of Bayesian differential privacy

Contact Info

Product

Resources

About