On the Security of Noise Addition for Privacy in Statistical Databases

Domingo‐Ferrer, Josep; Sebé, Francesc; Castellà‐Roca, Jordi

doi:10.1007/978-3-540-25955-8_12

Cited by 65 publications

(41 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multiway transportation problems form a very important class of discrete optimization problems and have been used and studied extensively in the operations research and mathematical programming literature, as well as in the statistics literature in the context of secure statistical data disclosure and management by public agencies, see e.g. [4,6,11,18,19,42,43,53,60,62] and the references therein. The feasible points in a transportation problem are the multiway tables ("contingency tables" in statistics) such that the sums of entries over some of their lower dimensional sub-tables such as lines or planes ("margins" in statistics) are specified.…”

Section: Corollary 415mentioning

confidence: 99%

Convex Discrete Optimization

Onn

2008

Encyclopedia of Optimization

View full text Add to dashboard Cite

We develop an algorithmic theory of convex optimization over discrete sets. Using a combination of algebraic and geometric tools we are able to provide polynomial time algorithms for solving broad classes of convex combinatorial optimization problems and convex integer programming problems in variable dimension. We discuss some of the many applications of this theory including to quadratic programming, matroids, bin packing and cutting-stock problems, vector partitioning and clustering, multiway transportation problems, and privacy and confidential statistical data disclosure. Highlights of our work include a strongly polynomial time algorithm for convex and linear combinatorial optimization over any family presented by a membership oracle when the underlying polytope has few edge-directions; a new theory of so-termed n-fold integer programming, yielding polynomial time solution of important and natural classes of convex and linear integer programming problems in variable dimension; and a complete complexity classification of high dimensional transportation problems, with practical applications to fundamental problems in privacy and confidential statistical data disclosure.

show abstract

Section: Corollary 415mentioning

confidence: 99%

Convex Discrete Optimization

Onn

2008

Encyclopedia of Optimization

View full text Add to dashboard Cite

show abstract

Section: Synthetic Treatments Of Microdatamentioning

confidence: 99%

“…In addition, every 2 years the UNESCO sponsors an international conference (i.e., Privacy in Statistical Databases) that gathers worldwide experts from different disciplines to discuss current issues in the field. Proceedings are published by Springer in the series Lecture Notes in Computer Science (Domingo-Ferrer, 2002;Domingo-Ferrer & Franconi, 2006;Domingo-Ferrer & Magkos, 2010;Domingo-Ferrer & Saygin, 2008;DomingoFerrer & Torra, 2004;European Communities, 1998).…”

Section: Need For Protection Need For Informationmentioning

confidence: 99%

“…A potential difficulty with this approach is that some reidentification experts suggest that knowing information about how the noise was applied to the variables can allow an intruder to reverse "engineer" the data file and potentially identify individuals. Details on how noise can be reverse engineered are available in Domingo-Ferrer, Sebe, and Castella-Roca (2004).Other forms of synthetic perturbation include data swapping or substitution for all records in the data set. This is an extension of the previously mentioned nonsynthetic perturbation methods, where selection for treatment is expanded from deterministic or random selection processes to 100% selection as part of a synthetic treatment.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Avoiding Disclosure of Individually Identifiable Health Information

et al. 2011

View full text Add to dashboard Cite

Achieving data and information dissemination without harming anyone is a central task of any entity in charge of collecting data. In this article, the authors examine the literature on data and statistical confidentiality. Rather than comparing the theoretical properties of specific methods, they emphasize the main themes that emerge from the ongoing discussion among scientists regarding how best to achieve the appropriate balance between data protection, data utility, and data dissemination. They cover the literature on de-identification and reidentification methods with emphasis on health care data. The authors also discuss the benefits and limitations for the most common access methods. Although there is abundant theoretical and empirical research, their review reveals lack of consensus on fundamental questions for empirical practice: How to assess disclosure risk, how to choose among disclosure methods, how to assess reidentification risk, and how to measure utility loss.Keywords public use files, disclosure avoidance, reidentification, de-identification, data utility 2 SAGE Open inferential disclosure (i.e., information that can be inferred about a record in a data set with better accuracy). There is significant literature on each of these topics, which are beyond the scope of this article.Our article is divided into six sections, of which this "Introduction" is the first. The second section presents "The Policy and Academic Context" surrounding the discussion. The third section discusses the state of the art in "De-Identification Methods," while the fourth emphasizes the state of the art in "Reidentification Methods." The fifth section presents the conclusions from the literature on the different ways in which users may "Access" public data, stressing the trade-offs between (a) confidentiality and utility and (b) confidentiality and ease of access. The last section presents the "Conclusion." The Policy and Academic Context Historic PerspectiveConcerns about privacy and confidentiality in governmental efforts to collect and disseminate information are not new. As a review by Anderson and Seltzer (2009) suggests, "the roots of the modern concept of federal statistical confidentiality can be traced directly back to the late nineteenth century" (p. 8). Notwithstanding this history, the literature on statistical disclosure methods is fairly recent by modern standards (Dalenius, 1977, is considered the seminal paper). In 1975, the U.S. Federal Committee on Statistical Methodology (FCSM) was organized by the Office of Management and Budget (OMB) to investigate issues of data quality affecting federal statistics. As part of this effort, the Subcommittee on Disclosure Limitation Methodology, created within the FCSM, published its 1994 Statistical Policy Working Paper 22 (SPWP22). This paper, which was revised in 2005 by the Confidentiality and Data Access Committee (CDAC, 2005), sets good practice guidelines and recommendations for all agencies regarding confidentiality protection. Defining Confidentiality ...

show abstract

“…However, when the covariance matrix of noise is proportional to the covariance matrix of the original data, the additive noise method can preserve the correlation coefficients (Domingo-Ferrer et al, 2004) and preserve regression coefficients. Due to the restriction on the covariance matrix of noise, the additive noise method will lead to asymptotically biased regression estimates for subpopulations; see Tendick and Norman (1987); Domingo-Ferrer et al (2004). Ting et al (2008) introduced the Random Orthogonal Matrix Masking (ROMM) procedure for perturbing data in regression analysis.…”

mentioning

confidence: 99%

Estimation of Regression Parameters from Noise Multiplied Data

Lin

Wise

2013

JPC

View full text Add to dashboard Cite

Abstract. This paper considers the scenario that all data entries in a confidentialised unit record file are masked by multiplicative noises, regardless of whether unit records are sensitive or not and regardless of whether the masked variables are dependent or independent variables in the underlying regression analysis. A technique is introduced in this paper to show how to estimate parameters in a regression model, which is originally fitted by unmasked data, based on masked data. Several simulation studies and a real-life data application are presented.

show abstract

On the Security of Noise Addition for Privacy in Statistical Databases

Cited by 65 publications

References 7 publications

Convex Discrete Optimization

Convex Discrete Optimization

Avoiding Disclosure of Individually Identifiable Health Information

Estimation of Regression Parameters from Noise Multiplied Data

Contact Info

Product

Resources

About