Indistinguishability: The Other Aspect of Privacy

Yao, Chao; Wang, Lingyu; Wang, Sean X.; Jajodia, Sushil

doi:10.1007/11844662_1

Cited by 12 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The idea of exploiting symmetries among sensitive values, originally proposed in the preliminary version of this paper [31] seems to have influenced another recent proposal of anonymization techniques. Indeed, Koudas et al [19] considered the problem of anonymizing data in order to answer aggregate queries and proposed a technique based on the permutation of sensitive attribute values to achieve anonymization, as an alternative to generalization.…”

Section: Related Workmentioning

confidence: 99%

Evaluating privacy threats in released database views by symmetric indistinguishability

Yao

Wang

et al. 2009

JCS

Self Cite

View full text Add to dashboard Cite

A privacy violation occurs when the association between an individual identity and data considered private by that individual is obtained by an unauthorized party. Uncertainty and indistinguishability are two independent aspects that characterize the degree of this association being revealed. Indistinguishability refers to the property that the attacker cannot see the difference among a group of individuals, while uncertainty refers to the property that the attacker cannot tell which private value, among a group of values, an individual actually has. This paper investigates the notion of indistinguishability as a general form of anonymity, applicable, for example, not only to generalized private tables, but to relational views and to sets of views obtained by multiple queries over a private database table. It is shown how indistinguishability is highly influenced by certain symmetries among individuals, in the released data, with respect to their private values. The paper provides both theoretical results and practical algorithms for checking if a specific set of views over a private table provide sufficient indistinguishability.

show abstract

Section: Related Workmentioning

confidence: 99%

Evaluating privacy threats in released database views by symmetric indistinguishability

Yao

Wang

et al. 2009

JCS

Self Cite

View full text Add to dashboard Cite

show abstract

“…Two important aspects of any privacy technique are uncertainty and indistinuishability [38,37]. Indistinuishability is defined as the inability of telling the difference among individuals in a group.…”

Section: Uncertainty and Indistinguishabilitymentioning

confidence: 99%

“…Indistinguishability. More strict notions (such as polynomial indistinguishability) are often used in cryptography, but in the database literature more practical metrics are usually applied, such as symmetric indistinguishability [38,37], defined next. Note that we do not publish T but publish both M () and its result on T , namely M (T ).…”

Section: Uncertainty and Indistinguishabilitymentioning

confidence: 99%

Publishing naive Bayesian classifiers

Mozafari

Zaniolo

2009

Proc. VLDB Endow.

View full text Add to dashboard Cite

We address the problem of publishing a Naïve Bayesian Classifier (NBC) or, equivalently, publishing the necessary views for building an NBC, while protecting privacy of the individuals who provided the training data. Our approach completely preserves the accuracy of the original classifier, and thus significantly improves on current approaches, such as randomization or anonymization, which typically degrade accuracy to preserve privacy. Current query-view security checkers address the question of 'Is the view safe to publish?' and are computationally expensive (often Π p 2 -complete). Here instead, we tackle the question of 'How to make a view safe to publish?' and propose a linear-time algorithm to publish safe NBCenabling views.We first show that a simple measure that restricts the ratios between the published NBC statistics is sufficient to prevent any breach of privacy. Then, we propose a linear-time algorithm to enforce this measure by producing perturbed statistics that assure both (i) individuals' privacy, and (ii) a classifier that behaves in the same way as the NBC trained on the original data. By carefully expressing the derived statistics using rational numbers, we can easily produce synthetic (sanitized) datasets. Thus, for any given dataset, we produce another dataset that is secure to publish (w.r.t. a uniform prior) and achieves the same classification accuracy. Finally, we extend our results by providing sufficient conditions to cope with arbitrary (non-uniform prior) distributions, and we validate their effectiveness in practice through experiments on real-world data.

show abstract

“…The current privacypreserving researches [3][4][5][6][7][8][9][10] focus on the protection of the single-owner privacy data, which is related to personal things, such as one's salary, preference and so on. Usually, it is easy to store and use such privacy data in relational data schema.…”

Section: Introductionmentioning

confidence: 99%

A privacy policy conflict detection method for multi-owner privacy data protection

Ren

Cheng

Peng

et al. 2010

Electron Commer Res

View full text Add to dashboard Cite

The current privacy-preserving researches focus on single-owner privacy data. However, multi-owner privacy data, which is also a widespread privacy data, need to be properly protected. At first, the characteristics of multi-owner privacy data and its protection requirement is introduced in this paper. Secondly, a data schema based on deputy mechanism for multi-owner privacy data is proposed. Thirdly, based on the schema, this paper proposes a privacy policy conflict detection method based on sub-graph isomorphic. This method models the privacy policy and each possible policy conflict pattern as a stratified-directed graph (SDG), and provides an algorithm to detect whether the SDG of a privacy conflict mode is isomorphic to that of privacy policies.Y. Ren ( ) The State-Key

show abstract

Indistinguishability: The Other Aspect of Privacy

Cited by 12 publications

References 29 publications

Evaluating privacy threats in released database views by symmetric indistinguishability

Evaluating privacy threats in released database views by symmetric indistinguishability

Publishing naive Bayesian classifiers

A privacy policy conflict detection method for multi-owner privacy data protection

Contact Info

Product

Resources

About