Random orthogonal matrix masking methodology for microdata release

Ting, Daniel; Fienberg, Stephen E.; Trottini, Mario

doi:10.1504/ijics.2008.016823

Cited by 26 publications

(28 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, in Mera (1997), symmetric orthogonal matrices were used in order to define moment‐preserving masks. The random orthogonal matrix masking (ROMM) method, introduced in Ting et al (2008), uses random orthogonal matrices. ROMM encompasses several perturbation methods.…”

Section: Statistical Disclosure Limitation Methodsmentioning

confidence: 99%

Disclosure Control of Business Microdata: A Density‐Based Approach

Ichim¹

2009

Int Statistical Rev

View full text Add to dashboard Cite

For continuous key variables, a measure of the individual risk of disclosure is proposed. This risk measure, the local outlier factor, estimates the density around a unit. A selective masking method based on the nearest-neighbour principle and microaggregation is also introduced. Some results of an application to the Italian sample of the Community Innovation Survey are presented. Copyright (c) 2009 The Author. Journal compilation (c) 2009 International Statistical Institute.

show abstract

Section: Statistical Disclosure Limitation Methodsmentioning

confidence: 99%

Disclosure Control of Business Microdata: A Density‐Based Approach

Ichim¹

2009

Int Statistical Rev

View full text Add to dashboard Cite

show abstract

“…These can be continuous, discrete, or a mixture of the two, and can even be objects such as pictures. We could apply any standard approach to protecting the release of information from such an array, such as the many variants on matrix masking [4], e.g., data perturbation, or more elaborate methods such as Random Orthogonal matrix manipulation [12], as well as invoking any of the usual criteria such as k-anonymity, l-diversity, differential privacy, etc. All of these methods typically take the rows of the array corresponding to individuals as realizations of independent random variables.…”

Section: Protecting the Privacy Of Network Datamentioning

confidence: 99%

Is the Privacy of Network Data an Oxymoron?

Fienberg

2013

JPC

Self Cite

View full text Add to dashboard Cite

While social networks are now a part of everyday life for the vast majority of people using computers, smartphones, and tablets, privacy is but an afterthought. Google+ has in excess of 100 million users a month while Facebook has topped 1 billion. Other more specialized networks such as Linked-in add to the fray. But from a privacy perspective the biggest concern for users should be the efforts to integrate the networking apps into all other forms of online activity as well as the constant effort to link additional data to network information, in addition to the network owners’ efforts to market that information to third party vendors. Further, Facebook and other networking sites have already begun to build search capabilities and other facilities into their systems that extend the information they collect from users even further. What protections do users really have?

show abstract

“…In the statistical disclosure limitation literature, Ting et al (2008) focus on data release of multivariate data and propose a multiplicative perturbation that preserves means and variance/covariances, and thus least squares regression coefficients and compares their results with approaches involving additive perturbation (Burridge, 2003;Muralidhar and Sarathy, 2006). O'Keefe and Good (2009) consider the release of regression results via remote analysis servers.…”

Section: Overview Of Related Literaturementioning

confidence: 99%

Privacy-Preserving Data Sharing in High Dimensional Regression and Classification Settings

Fienberg

Jin

2012

JPC

Self Cite

View full text Add to dashboard Cite

Abstract. We focus on the problem of multi-party data sharing in high dimensional data settings where the number of measured features (or the dimension) p is frequently much larger than the number of subjects (or the sample size) n, the so-called p n scenario that has been the focus of much recent statistical research. Here, we consider data sharing for two interconnected problems in high dimensional data analysis, namely the feature selection and classification. We characterize the notions of "cautious", "regular", and "generous" data sharing in terms of their privacy-preserving implications for the parties and their share of data, with focus on the "feature privacy" rather than the "sample privacy," though the violation of the former may lead to the latter. We evaluate the data sharing methods using a phase diagram from the statistical literature on multiplicity and Higher Criticism thresholding. In the two-dimensional phase space calibrated by the signal sparsity and signal strength, a phase diagram is a partition of the phase space and contains three distinguished regions, where we have no (feature) privacy violation, relatively rare privacy violations, and an overwhelming amount of privacy violation.

show abstract

Random orthogonal matrix masking methodology for microdata release

Cited by 26 publications

References 17 publications

Disclosure Control of Business Microdata: A Density‐Based Approach

Disclosure Control of Business Microdata: A Density‐Based Approach

Is the Privacy of Network Data an Oxymoron?

Privacy-Preserving Data Sharing in High Dimensional Regression and Classification Settings

Contact Info

Product

Resources

About