Survey on Improving Data Utility in Differentially Private Sequential Data Publishing

Yang, Xinyu; Wang, Teng; Ren, Xuebin; Yu, Wei

doi:10.1109/tbdata.2017.2715334

Cited by 71 publications

(39 citation statements)

References 112 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since LDP no longer assumes a trusted third-party data curator, the interactive and non-interactive privacy model settings of LDP [ 47 ] are different from that of DP [ 23 ]. Figure 3 shows the interactive and non-interactive settings of LDP.…”

Section: Theoretical Summarization Of Ldpmentioning

confidence: 99%

“…(Without loss of generality, DP appears in the rest of this article refers to global DP (i.e., centralized DP).) With strict mathematical proofs, DP is independent of the background knowledge of adversaries and capable of providing each user with strong privacy guarantees, which was widely adopted and used in many areas [ 23 , 24 ]. However, DP can be only used to the assumption of a trusted server.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Comprehensive Survey on Local Differential Privacy toward Data Statistics and Analysis

Wang

Zhang

Feng

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Collecting and analyzing massive data generated from smart devices have become increasingly pervasive in crowdsensing, which are the building blocks for data-driven decision-making. However, extensive statistics and analysis of such data will seriously threaten the privacy of participating users. Local differential privacy (LDP) was proposed as an excellent and prevalent privacy model with distributed architecture, which can provide strong privacy guarantees for each user while collecting and analyzing data. LDP ensures that each user’s data is locally perturbed first in the client-side and then sent to the server-side, thereby protecting data from privacy leaks on both the client-side and server-side. This survey presents a comprehensive and systematic overview of LDP with respect to privacy models, research tasks, enabling mechanisms, and various applications. Specifically, we first provide a theoretical summarization of LDP, including the LDP model, the variants of LDP, and the basic framework of LDP algorithms. Then, we investigate and compare the diverse LDP mechanisms for various data statistics and analysis tasks from the perspectives of frequency estimation, mean estimation, and machine learning. Furthermore, we also summarize practical LDP-based application scenarios. Finally, we outline several future research directions under LDP.

show abstract

Section: Theoretical Summarization Of Ldpmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Comprehensive Survey on Local Differential Privacy toward Data Statistics and Analysis

Wang

Zhang

Feng

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Data Privacy is the appropriate use of data available with any individual or any organization, unlike data security that guarantees confidentiality, integrity, availability of data. Some of the well known data privacy preservation methods are k-anonymization [1] [2], l-diversity [2][3], t-closeness [4] and differential privacy [5]. Table 1 summarizes these algorithms along with their pros and cons.…”

Section: Introductionmentioning

confidence: 99%

Machine Learning Concepts for Correlated Big Data Privacy

Biswas¹,

Khare²,

Agrawal³

et al. 2021

Preprint

View full text Add to dashboard Cite

With data becoming a salient asset worldwide, dependence within data kept on growing, hence the real world datasets that one works upon in today's time are highly correlated. Since the past few years, researchers have given attention to this aspect of data privacy and found that where there exists a correlation among data, the existing privacy guarantees could not be assured with existing privacy algorithms. The privacy guarantees provided by existing algorithms were enough when there existed no relation between data in the datasets. Hence, by keeping the existence of data correlation into account, there is a dire need, to reconsider the privacy algorithms. Some of the research have considered to utilize a well known machine learning concept, i.e., Data Correlation Analysis to understand the relationship between data in a better way. This has given some promising results as well. Though its less but still a considerable amount of research has been done for correlated data privacy. But correlated big data privacy is very less explored. The real world datasets that are worked upon, are often large in size (technologically termed as big data) and house a high amount of data correlation. Hence, there is a grave need to understand and propose solutions for correlated big data privacy.

show abstract

“…The emphasis on privacy issues has promoted research on privacy protection technologies, for which the degree of privacy protection and the utility of data are the most important metrics. In line with this developmental trend, differential privacy technology [2], [3] has been proposed. As a privacy protection model, it strictly defines the strength of privacy protection, that is, the addition or deletion of any record will not affect the final query result.…”

Section: Introductionmentioning

confidence: 99%

Combinational Randomized Response Mechanism for Unbalanced Multivariate Nominal Attributes

Feng

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

This work was supported by "the Construction team project of the introduction and cultivation of young innovative talents in Colleges and universities of Shandong Province of China, 2019. (Project name: Big data and business intelligence social service innovation team)" ABSTRACT At present, many enterprises provide users with better services by collecting their sensitive information. However, these enterprises will inevitably cause the leakage of users' information, thereby infringing on users' privacy. Local differential privacy resolves this problem by only aggregating randomized values from each user, providing plausible deniability. However, different users might have diverse privacy requirements for different attributes. Moreover, the dimensions of these attributes may be unbalanced. Traditional local differential privacy algorithms usually assign the same privacy budget to all attributes, resulting in undesired frequency estimation. To obtain highly accurate of the results while satisfying local differential privacy, the aggregator needs to implement a reasonable privacy budget allocation scheme. Motivated by this, this paper proposed a novel local differential privacy scheme. The proposed method combines the advantages of BRR and MRR to address the problem of high and low privacy requirements. It employs the Lagrange multiplier algorithm to transform the privacy budget allocation problem between unbalanced attributes into a problem of calculating minima from unconditionally constrained convex functions. The solution to the resulting nonlinear equation is used as the final privacy budget allocation scheme. Simulation experiments show that the novel local differential privacy scheme proposed by this paper can significantly reduce the estimation error under the premise of satisfying the local differential privacy.

show abstract

Survey on Improving Data Utility in Differentially Private Sequential Data Publishing

Cited by 71 publications

References 112 publications

A Comprehensive Survey on Local Differential Privacy toward Data Statistics and Analysis

A Comprehensive Survey on Local Differential Privacy toward Data Statistics and Analysis

Machine Learning Concepts for Correlated Big Data Privacy

Combinational Randomized Response Mechanism for Unbalanced Multivariate Nominal Attributes

Contact Info

Product

Resources

About