Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis

Raúl Pardo; Willard Rafnsson; Christian Probst; Andrzej Wasowski

doi:10.1007/978-3-030-88428-4_21

Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis

UNITEC Institute of Technology

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

Disclosure of data analytics results has important scientific and commercial justifications. However, no data shall be disclosed without a diligent investigation of risks for privacy of subjects. Privug is a tool-supported method to explore information leakage properties of data analytics and anonymization programs. In Privug, we reinterpret a program probabilistically, using off-the-shelf tools for Bayesian inference to perform information-theoretic analysis of the information flow. For privacy researchers, Privug provides a fast, lightweight way to experiment with privacy protection measures and mechanisms. We show that Privug is accurate, scalable, and applicable to a range of leakage analysis scenarios.

Original language	English
Title of host publication	European Symposium on Research in Computer Security : Computer Security – ESORICS 2021
Volume	12973
Publisher	Springer
Publication date	2021
ISBN (Print)	978-3-030-88427-7
ISBN (Electronic)	978-3-030-88428-4
DOIs	https://doi.org/10.1007/978-3-030-88428-4_21
Publication status	Published - 2021
Event	European Symposium on Research in Computer Security - VIRTUAL Duration: 4 Oct 2021 → 8 Oct 2021 Conference number: 26

Conference

Conference	European Symposium on Research in Computer Security
Number	26
City	VIRTUAL
Period	04/10/2021 → 08/10/2021

Series	Lecture Notes in Computer Science
Volume	12973
ISSN	0302-9743

Keywords

Data Analytics
Privacy Protection
Information Leakage
Anonymization Programs
Bayesian Inference

Access to Document

10.1007/978-3-030-88428-4_21

https://pure.itu.dk/admin/files/93413939/21_esorics_privug.pdfSubmitted manuscript, 1.19 MBLicence: Unspecified

Cite this

@inproceedings{95da214d830941ba88cddcbb7fd4b2f1,

title = "Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis",

abstract = "Disclosure of data analytics results has important scientific and commercial justifications. However, no data shall be disclosed without a diligent investigation of risks for privacy of subjects. Privug is a tool-supported method to explore information leakage properties of data analytics and anonymization programs. In Privug, we reinterpret a program probabilistically, using off-the-shelf tools for Bayesian inference to perform information-theoretic analysis of the information flow. For privacy researchers, Privug provides a fast, lightweight way to experiment with privacy protection measures and mechanisms. We show that Privug is accurate, scalable, and applicable to a range of leakage analysis scenarios.",

keywords = "Data Analytics, Privacy Protection, Information Leakage, Anonymization Programs, Bayesian Inference, Data Analytics, Privacy Protection, Information Leakage, Anonymization Programs, Bayesian Inference",

author = "Ra{\'u}l Pardo and Willard Rafnsson and Christian Probst and Andrzej Wasowski",

year = "2021",

doi = "10.1007/978-3-030-88428-4\_21",

language = "English",

isbn = "978-3-030-88427-7",

volume = "12973",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

booktitle = "European Symposium on Research in Computer Security",

address = "Germany",

note = "European Symposium on Research in Computer Security ; Conference date: 04-10-2021 Through 08-10-2021",

}

Pardo, R , Rafnsson, W, Probst, C & Wasowski, A 2021, Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis. in European Symposium on Research in Computer Security: Computer Security – ESORICS 2021. vol. 12973, Springer, Lecture Notes in Computer Science, vol. 12973, European Symposium on Research in Computer Security, VIRTUAL, 04/10/2021. https://doi.org/10.1007/978-3-030-88428-4_21

Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis. / Pardo, Raúl ; Rafnsson, Willard; Probst, Christian et al.
European Symposium on Research in Computer Security: Computer Security – ESORICS 2021. Vol. 12973 Springer, 2021. (Lecture Notes in Computer Science, Vol. 12973).

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis

AU - Pardo, Raúl

AU - Rafnsson, Willard

AU - Probst, Christian

AU - Wasowski, Andrzej

N1 - Conference code: 26

PY - 2021

Y1 - 2021

N2 - Disclosure of data analytics results has important scientific and commercial justifications. However, no data shall be disclosed without a diligent investigation of risks for privacy of subjects. Privug is a tool-supported method to explore information leakage properties of data analytics and anonymization programs. In Privug, we reinterpret a program probabilistically, using off-the-shelf tools for Bayesian inference to perform information-theoretic analysis of the information flow. For privacy researchers, Privug provides a fast, lightweight way to experiment with privacy protection measures and mechanisms. We show that Privug is accurate, scalable, and applicable to a range of leakage analysis scenarios.

AB - Disclosure of data analytics results has important scientific and commercial justifications. However, no data shall be disclosed without a diligent investigation of risks for privacy of subjects. Privug is a tool-supported method to explore information leakage properties of data analytics and anonymization programs. In Privug, we reinterpret a program probabilistically, using off-the-shelf tools for Bayesian inference to perform information-theoretic analysis of the information flow. For privacy researchers, Privug provides a fast, lightweight way to experiment with privacy protection measures and mechanisms. We show that Privug is accurate, scalable, and applicable to a range of leakage analysis scenarios.

KW - Data Analytics

KW - Privacy Protection

KW - Information Leakage

KW - Anonymization Programs

KW - Bayesian Inference

KW - Data Analytics

KW - Privacy Protection

KW - Information Leakage

KW - Anonymization Programs

KW - Bayesian Inference

U2 - 10.1007/978-3-030-88428-4_21

DO - 10.1007/978-3-030-88428-4_21

M3 - Article in proceedings

SN - 978-3-030-88427-7

VL - 12973

T3 - Lecture Notes in Computer Science

BT - European Symposium on Research in Computer Security

PB - Springer

T2 - European Symposium on Research in Computer Security

Y2 - 4 October 2021 through 8 October 2021

ER -

Privug: Using Probabilistic Programming for Quantifying Leakage in Privacy Risk Analysis

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this