Local synthesis for disclosure limitation that satisfies probabilistic k-anonymity criterion

Oganian, A.; Domingo-Ferrer, J.

Belongs to PC:SerieArticles collection

TITLE:

Local synthesis for disclosure limitation that satisfies probabilistic k-anonymity criterion - PC:2893

Handle:

https://hdl.handle.net/20.500.11797/PC2893

URV's Author/s:	OGANIAN , ANNA; DOMINGO FERRER, JOSEP
Author, as appears in the article.:	Oganian, A.; Domingo-Ferrer, J.
Author identifier:	; 0000-0001-7213-4962
Journal publication year:	2017
Publication Type:	Article
ISSN:	1888-5063
Abstract:	Before releasing databases which contain sensitive information about individuals, data publishers must apply Statistical Disclosure Limitation (SDL) methods to them, in order to avoid disclosure of sensitive information on any identifiable data subject. SDL methods often consist of masking or synthesizing the original data records in such a way as to minimize the risk of disclosure of the sensitive information while providing data users with accurate information about the population of interest. In this paper we propose a new scheme for disclosure limitation, based on the idea of local synthesis of data. Our approach is predicated on model-based clustering. The proposed method satisfies the requirements of k-anonymity; in particular we use a variant of the k-anonymity privacy model, namely probabilistic k-anonymity, by incorporating constraints on cluster cardinality. Regarding data utility, for continuous attributes, we exactly preserve means and covariances of the original data, while approximately preserving higher-order moments and analyses on subdomains (defined by clusters and cluster combinations). For both continuous and categorical data, our experiments with medical data sets show that, from the point of view of data utility, local synthesis compares very favorably with other methods of disclosure limitation including the sequential regression approach for synthetic data generation.
Link to the original source:	http://www.tdp.cat/issues16/vol10n01.php
Papper version:	info:eu-repo/semantics/publishedVersion
licence for use:	https://creativecommons.org/licenses/by/3.0/es/
Department:	Enginyeria Informàtica i Matemàtiques
Research group:	Seguretat i Privadesa
Licence document URL:	https://repositori.urv.cat/ca/proteccio-de-dades/
Thematic Areas:	Mathematics
Keywords:	Probabilistic k-anonymity Mixture model Expectation-Maximization (EM) algorithm
Awards and grants:	ICREA Acadèmia
Entity:	Universitat Rovira i Virgili
Record's date:	2017-05-26
First page:	61
Last page:	81
Journal volume:	10

Description:	Filiació URV: SI DOI: N/D URL: http://www.tdp.cat/issues16/vol10n01.php Before releasing databases which contain sensitive information about individuals, data publishers must apply Statistical Disclosure Limitation (SDL) methods to them, in order to avoid disclosure of sensitive information on any identifiable data subject. SDL methods often consist of masking or synthesizing the original data records in such a way as to minimize the risk of disclosure of the sensitive information while providing data users with accurate information about the population of interest. In this paper we propose a new scheme for disclosure limitation, based on the idea of local synthesis of data. Our approach is predicated on model-based clustering. The proposed method satisfies the requirements of k-anonymity; in particular we use a variant of the k-anonymity privacy model, namely probabilistic k-anonymity, by incorporating constraints on cluster cardinality. Regarding data utility, for continuous attributes, we exactly preserve means and covariances of the original data, while approximately preserving higher-order moments and analyses on subdomains (defined by clusters and cluster combinations). For both continuous and categorical data, our experiments with medical data sets show that, from the point of view of data utility, local synthesis compares very favorably with other methods of disclosure limitation including the sequential regression approach for synthetic data generation.
Coverage:	Anglès
Type:	article Article Artículo Article info:eu-repo/semantics/publishedVersion
Contributor:	Seguretat i Privadesa Enginyeria Informàtica i Matemàtiques Universitat Rovira i Virgili
Títol:	Local synthesis for disclosure limitation that satisfies probabilistic k-anonymity criterion
Subject:	Xifratge (Informàtica) Protecció de dades Probabilistic k-anonymity Mixture model Expectation-Maximization (EM) algorithm Mathematics Matemáticas Matemàtiques 1888-5063
Date:	2017
Format:	311 kb
Creator:	Oganian, A. Domingo-Ferrer, J.
Rights:	info:eu-repo/semantics/openAccess
Publisher:	Publicacions de la Universitat Autònoma de Barcelona

Search your record at:

Available files
File	Description	Format
DocumentPrincipal	DocumentPrincipal	application/pdf

All objects of this collection

Information