Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata

Claire Little; Mark Elliot; Richard Allmendinger

doi:10.1007/978-3-031-13945-1_17

Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata

Claire Little, Mark Elliot, Richard Allmendinger

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Most statistical agencies release randomly selected samples of Census microdata, usually with sample fractions under 10% and with other forms of statistical disclosure control (SDC) applied. An alternative to SDC is data synthesis, which has been attracting growing interest, yet there is no clear consensus on how to measure the associated utility and disclosure risk of the data. The ability to produce synthetic Census microdata, where the utility and associated risks are clearly understood, could mean that more timely and wider-ranging access to microdata would be possible. This paper follows on from previous work by the authors which mapped synthetic Census data on a risk-utility (R-U) map. The paper presents a framework to measure the utility and disclosure risk of synthetic data by comparing it to samples of the original data of varying sample fractions, thereby identifying the sample fraction which has equivalent utility and risk to the synthetic data. Three commonly used data synthesis packages are compared with some interesting results. Further work is needed in several directions but the methodology looks very promising.

Original language	English
Title of host publication	Privacy in Statistical Databases
Subtitle of host publication	International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings
Editors	Josep Domingo-Ferrer, Maryline Laurent
Place of Publication	Cham, Switzerland
Publisher	Springer Nature
Pages	234-249
Number of pages	16
ISBN (Electronic)	9783031139451
ISBN (Print)	9783031139444
DOIs	https://doi.org/10.1007/978-3-031-13945-1_17
Publication status	Published - 14 Sep 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13463 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Keywords

data synthesis
data utility
disclosure risk
microdata

Research Beacons, Institutes and Platforms

Cathie Marsh Institute

Access to Document

10.1007/978-3-031-13945-1_17

Cite this

Little, C., Elliot, M., & Allmendinger, R. (2022). Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata. In J. Domingo-Ferrer, & M. Laurent (Eds.), Privacy in Statistical Databases: International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings (pp. 234-249). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13463 LNCS). Springer Nature. https://doi.org/10.1007/978-3-031-13945-1_17

Little, Claire ; Elliot, Mark ; Allmendinger, Richard. / Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata. Privacy in Statistical Databases: International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings. editor / Josep Domingo-Ferrer ; Maryline Laurent. Cham, Switzerland : Springer Nature, 2022. pp. 234-249 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c56f6a46cf064405b8e4e515e6fd196c,

title = "Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata",

abstract = "Most statistical agencies release randomly selected samples of Census microdata, usually with sample fractions under 10% and with other forms of statistical disclosure control (SDC) applied. An alternative to SDC is data synthesis, which has been attracting growing interest, yet there is no clear consensus on how to measure the associated utility and disclosure risk of the data. The ability to produce synthetic Census microdata, where the utility and associated risks are clearly understood, could mean that more timely and wider-ranging access to microdata would be possible. This paper follows on from previous work by the authors which mapped synthetic Census data on a risk-utility (R-U) map. The paper presents a framework to measure the utility and disclosure risk of synthetic data by comparing it to samples of the original data of varying sample fractions, thereby identifying the sample fraction which has equivalent utility and risk to the synthetic data. Three commonly used data synthesis packages are compared with some interesting results. Further work is needed in several directions but the methodology looks very promising.",

keywords = "data synthesis, data utility, disclosure risk, microdata",

author = "Claire Little and Mark Elliot and Richard Allmendinger",

note = "Funding Information: The authors wish to acknowledge IPUMs International and the statistical offices that provided the underlying data making this research possible: Statistics Canada; Bureau of Statistics, Fiji; National Institute of Statistics, Rwanda; and the Office for National Statistics, UK. Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.",

year = "2022",

month = sep,

day = "14",

doi = "10.1007/978-3-031-13945-1_17",

language = "English",

isbn = "9783031139444",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Nature",

pages = "234--249",

editor = "Josep Domingo-Ferrer and Maryline Laurent",

booktitle = "Privacy in Statistical Databases",

address = "United States",

}

Little, C , Elliot, M & Allmendinger, R 2022, Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata. in J Domingo-Ferrer & M Laurent (eds), Privacy in Statistical Databases: International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13463 LNCS, Springer Nature, Cham, Switzerland, pp. 234-249. https://doi.org/10.1007/978-3-031-13945-1_17

Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata. / Little, Claire ; Elliot, Mark ; Allmendinger, Richard.

Privacy in Statistical Databases: International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings. ed. / Josep Domingo-Ferrer; Maryline Laurent. Cham, Switzerland : Springer Nature, 2022. p. 234-249 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13463 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata

AU - Little, Claire

AU - Elliot, Mark

AU - Allmendinger, Richard

N1 - Funding Information: The authors wish to acknowledge IPUMs International and the statistical offices that provided the underlying data making this research possible: Statistics Canada; Bureau of Statistics, Fiji; National Institute of Statistics, Rwanda; and the Office for National Statistics, UK. Publisher Copyright: © 2022, Springer Nature Switzerland AG.

PY - 2022/9/14

Y1 - 2022/9/14

N2 - Most statistical agencies release randomly selected samples of Census microdata, usually with sample fractions under 10% and with other forms of statistical disclosure control (SDC) applied. An alternative to SDC is data synthesis, which has been attracting growing interest, yet there is no clear consensus on how to measure the associated utility and disclosure risk of the data. The ability to produce synthetic Census microdata, where the utility and associated risks are clearly understood, could mean that more timely and wider-ranging access to microdata would be possible. This paper follows on from previous work by the authors which mapped synthetic Census data on a risk-utility (R-U) map. The paper presents a framework to measure the utility and disclosure risk of synthetic data by comparing it to samples of the original data of varying sample fractions, thereby identifying the sample fraction which has equivalent utility and risk to the synthetic data. Three commonly used data synthesis packages are compared with some interesting results. Further work is needed in several directions but the methodology looks very promising.

AB - Most statistical agencies release randomly selected samples of Census microdata, usually with sample fractions under 10% and with other forms of statistical disclosure control (SDC) applied. An alternative to SDC is data synthesis, which has been attracting growing interest, yet there is no clear consensus on how to measure the associated utility and disclosure risk of the data. The ability to produce synthetic Census microdata, where the utility and associated risks are clearly understood, could mean that more timely and wider-ranging access to microdata would be possible. This paper follows on from previous work by the authors which mapped synthetic Census data on a risk-utility (R-U) map. The paper presents a framework to measure the utility and disclosure risk of synthetic data by comparing it to samples of the original data of varying sample fractions, thereby identifying the sample fraction which has equivalent utility and risk to the synthetic data. Three commonly used data synthesis packages are compared with some interesting results. Further work is needed in several directions but the methodology looks very promising.

KW - data synthesis

KW - data utility

KW - disclosure risk

KW - microdata

UR - http://www.scopus.com/inward/record.url?scp=85138824654&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/e3e7f02b-9eee-301e-8ad9-ad44b4e31913/

U2 - 10.1007/978-3-031-13945-1_17

DO - 10.1007/978-3-031-13945-1_17

M3 - Conference contribution

SN - 9783031139444

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 234

EP - 249

BT - Privacy in Statistical Databases

A2 - Domingo-Ferrer, Josep

A2 - Laurent, Maryline

PB - Springer Nature

CY - Cham, Switzerland

ER -

Little C , Elliot M , Allmendinger R. Comparing the Utility and Disclosure Risk of Synthetic Data with Samples of Microdata. In Domingo-Ferrer J, Laurent M, editors, Privacy in Statistical Databases: International Conference, PSD 2022, Paris, France, September 21–23, 2022, Proceedings. Cham, Switzerland: Springer Nature. 2022. p. 234-249. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-13945-1_17