Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

SafePub: A Truthful Data Anonymization Algorithm With Strong Privacy Guarantees

SafePub: A Truthful Data Anonymization Algorithm With Strong Privacy Guarantees AbstractMethods for privacy-preserving data publishing and analysis trade off privacy risks for individuals against the quality of output data. In this article, we present a data publishing algorithm that satisfies the differential privacy model. The transformations performed are truthful, which means that the algorithm does not perturb input data or generate synthetic output data. Instead, records are randomly drawn from the input dataset and the uniqueness of their features is reduced. This also offers an intuitive notion of privacy protection. Moreover, the approach is generic, as it can be parameterized with different objective functions to optimize its output towards different applications. We show this by integrating six well-known data quality models. We present an extensive analytical and experimental evaluation and a comparison with prior work. The results show that our algorithm is the first practical implementation of the described approach and that it can be used with reasonable privacy parameters resulting in high degrees of protection. Moreover, when parameterizing the generic method with an objective function quantifying the suitability of data for building statistical classifiers, we measured prediction accuracies that compare very well with results obtained using state-of-the-art differentially private classification algorithms. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Proceedings on Privacy Enhancing Technologies de Gruyter

SafePub: A Truthful Data Anonymization Algorithm With Strong Privacy Guarantees

Download PDF
21 pages

Loading next page...
 
/lp/de-gruyter/safepub-a-truthful-data-anonymization-algorithm-with-strong-privacy-AoltjkdXkl
Publisher
de Gruyter
Copyright
© 2018 Raffael Bild et al., published by De Gruyter Open
ISSN
2299-0984
eISSN
2299-0984
DOI
10.1515/popets-2018-0004
Publisher site
See Article on Publisher Site

Abstract

AbstractMethods for privacy-preserving data publishing and analysis trade off privacy risks for individuals against the quality of output data. In this article, we present a data publishing algorithm that satisfies the differential privacy model. The transformations performed are truthful, which means that the algorithm does not perturb input data or generate synthetic output data. Instead, records are randomly drawn from the input dataset and the uniqueness of their features is reduced. This also offers an intuitive notion of privacy protection. Moreover, the approach is generic, as it can be parameterized with different objective functions to optimize its output towards different applications. We show this by integrating six well-known data quality models. We present an extensive analytical and experimental evaluation and a comparison with prior work. The results show that our algorithm is the first practical implementation of the described approach and that it can be used with reasonable privacy parameters resulting in high degrees of protection. Moreover, when parameterizing the generic method with an objective function quantifying the suitability of data for building statistical classifiers, we measured prediction accuracies that compare very well with results obtained using state-of-the-art differentially private classification algorithms.

Journal

Proceedings on Privacy Enhancing Technologies – de Gruyter

Published: Jan 1, 2018

References

  • Mondrian multidimensional anonymity Conference on Data pages
    LeFevre
  • Hiding the presence of individuals from shared databases InInternational Conference on Management of Data pages
    Nergiz
  • The cost of privacy : Destruction of data - mining utility in anonymized data publishing SIGKDD International Conference on Knowledge Discovery and Data Mining pages
    Brickell
  • Incognito Efficient full - domain anonymity InInternational Conference on Management of Data pages
    LeFevre
  • Comas closeness through microaggregation : Strict privacy with enhanced utility on Knowledge and Data
    Soria
  • Introduction to Privacy - Preserving Data Publishing Concepts Techniques
    Fung
  • Foundation for publishing company nd edition
    Graham
  • private data release for data mining Conference on Knowledge Discovery and Data Mining pages
    Mohammed
  • de loss through global recoding and local suppression
    Waal
  • Mechanism design via differential privacy on Foundations of pages
    McSherry
  • Data mining with differential privacy InInternational Conference on Knowledge Discovery and Data Mining pages
    Friedman
  • Differential privacy machine learning a survey
    Lipton
  • Differential - private data publishing through component analysis Transactions on Data Privacy
    Jiang
  • globally optimal k - anonymity method for the de - identification of health data
    Emam
  • supermodularity - based differential privacy preserving algorithm for data anonymization on Knowledge and Data
    Fouad
  • Protecting privacy using anonymity
    Emam
  • Flash efficient stable optimal anonymity In Conference on Privacy Risk Trust PASSAT Conference on Social pages
    Kohlmayer
  • Prasser tool for optimizing de - identified health data for use in statistical classification International Symposium on Computer - Based Medical Systems
    InIEEE
  • Incognito Efficient full - domain anonymity InInternational Conference on Management of Data pages
    LeFevre
  • private data release for data mining Conference on Knowledge Discovery and Data Mining pages
    Mohammed
  • Data mining with differential privacy InInternational Conference on Knowledge Discovery and Data Mining pages
    Friedman
  • Appendix Concepts and methods for de - identifying clinical trial data InSharing Clinical Trial Data : Maximizing Benefits Minimizing Risk pages National Academies US
    El Emam
  • practical framework for privacy - preserving data analytics InInternational Conference on pages
    Fan
  • processing and machine learning with differential privacy and challenges for continuous data
    Sarwate
  • On sampling anonymization and differential privacy Or anonymization meets differential privacy on Communications pages
    Li
  • Differentially private naive bayes classification International Joint Conferences on Web Intelligence and Intelligent Agent Technologies pages
    Vaidya
  • Using feature selection to improve the utility of differentially private data publishing
    Jafer
  • machine learning repository http archive ics uci edu ml
    Lichman
  • Hiding the presence of individuals from shared databases InInternational Conference on Management of Data pages
    Nergiz
  • Prasser Lightning Utility - driven anonymization of high - dimensional data Transactions on Data Privacy
  • Workloadaware anonymization techniques for large - scale datasets on
    LeFevre
  • Protecting privacy using anonymity
    Emam
  • Privacy integrated queries : An extensible platform for privacy - preserving data analysis InInternational Conference on Management of Data pages
    McSherry
  • Workloadaware anonymization techniques for large - scale datasets on
    LeFevre
  • diversity Privacy beyond kanonymity Transactions on Knowledge Discovery from Data
    Machanavajjhala
  • On the tradeoff between privacy and utility in data publishing InInternational Conference on Knowledge Discovery and Data Mining pages
    Li
  • Achieving anonymity privacy protection using generalization and suppression of and Knowledge - Based
    Sweeney
  • Data Media
    Emam
  • The importance of context : Risk - based de - identification of biomedical data Methods of information in medicine
    Prasser
  • The cost of privacy : Destruction of data - mining utility in anonymized data publishing SIGKDD International Conference on Knowledge Discovery and Data Mining pages
    Brickell
  • private search log sanitization with optimal output utility Conference on Extending Database Technology pages
    Hong
  • private search log sanitization with optimal output utility Conference on Extending Database Technology pages
    Hong
  • The importance of context : Risk - based de - identification of biomedical data Methods of information in medicine
    Prasser
  • globally optimal k - anonymity method for the de - identification of health data
    Emam
  • Distributed data mining with differential privacy International Conference on Communications pages
    Zhang
  • Provably private data anonymization Or anonymity meets differential privacy
    Li
  • Data Media
    Emam
  • practical framework for privacy - preserving data analytics InInternational Conference on pages
    Fan
  • Transforming data to satisfy privacy constraints InInternational Conference on Knowledge Discovery and Data Mining pages
    Iyengar
  • Conservative or liberal personalized differential privacy InIEEE International Conference on Data pages
    Jorgensen
  • Differential - private data publishing through component analysis Transactions on Data Privacy
    Jiang
  • An ad omnia approach to defining and achieving private data analysis InInternational Conference on Privacy and Trust in pages
    Dwork
  • Mechanism design via differential privacy on Foundations of pages
    McSherry
  • Crowd blending privacy in pages
    Gehrke
  • Appendix Concepts and methods for de - identifying clinical trial data InSharing Clinical Trial Data : Maximizing Benefits Minimizing Risk pages National Academies US
    El Emam
  • Using feature selection to improve the utility of differentially private data publishing
    Jafer
  • Foundation for publishing company nd edition
    Graham
  • supermodularity - based differential privacy preserving algorithm for data anonymization on Knowledge and Data
    Fouad
  • Principles - versus rules - based output statistical disclosure control in remote access environments
    Ritchie
  • Comas closeness through microaggregation : Strict privacy with enhanced utility on Knowledge and Data
    Soria
  • de loss through global recoding and local suppression
    Waal
  • game theoretic framework for analyzing re - identification risk one
    Wan
  • Empirical privacy and empirical utility of anonymized data InIEEE International Conference on Data Engineering Workshops pages
    Cormode
  • Achieving anonymity privacy protection using generalization and suppression of and Knowledge - Based
    Sweeney
  • Mondrian multidimensional anonymity Conference on Data pages
    LeFevre
  • Statistical disclosure control in practice Media
    Willenborg
  • Crowd blending privacy in pages
    Gehrke
  • Introduction to Privacy - Preserving Data Publishing Concepts Techniques
    Fung
  • On sampling anonymization and differential privacy Or anonymization meets differential privacy on Communications pages
    Li
  • Prasser tool for optimizing de - identified health data for use in statistical classification International Symposium on Computer - Based Medical Systems
    InIEEE
  • game theoretic framework for analyzing re - identification risk one
    Wan
  • Practicing differential privacy in health care Transactions on Data Privacy
    Dankar
  • Non - interactive differential privacy survey InInternational Workshop on Open Data pages
    Leoni
  • machine learning repository http archive ics uci edu ml
    Lichman
  • Privacy integrated queries : An extensible platform for privacy - preserving data analysis InInternational Conference on Management of Data pages
    McSherry
  • Pass Towards privacy for social networks zero - knowledge based definition of privacy InTheory of Cryptography Conference pages
    Gehrke
  • Practicing differential privacy in health care Transactions on Data Privacy
    Dankar
  • Differential privacy machine learning a survey
    Lipton
  • Conservative or liberal personalized differential privacy InIEEE International Conference on Data pages
    Jorgensen
  • diversity Privacy beyond kanonymity Transactions on Knowledge Discovery from Data
    Machanavajjhala
  • Differentially private naive bayes classification International Joint Conferences on Web Intelligence and Intelligent Agent Technologies pages
    Vaidya
  • a system for evaluating and comparing relational and transaction anonymization algorithms InInternational Conference on Extending Database Technology pages
    Poulis
  • Eibe Data mining Practical machine learning tools and techniques
    Witten
  • Transforming data to satisfy privacy constraints InInternational Conference on Knowledge Discovery and Data Mining pages
    Iyengar
  • An ad omnia approach to defining and achieving private data analysis InInternational Conference on Privacy and Trust in pages
    Dwork
  • Eibe Data mining Practical machine learning tools and techniques
    Witten
  • Flash efficient stable optimal anonymity In Conference on Privacy Risk Trust PASSAT Conference on Social pages
    Kohlmayer
  • Towards a differentially private data anonymization Report
    Fouad
  • Provably private data anonymization Or anonymity meets differential privacy
    Li
  • Prasser Lightning Utility - driven anonymization of high - dimensional data Transactions on Data Privacy
  • Pass Towards privacy for social networks zero - knowledge based definition of privacy InTheory of Cryptography Conference pages
    Gehrke
  • Non - interactive differential privacy survey InInternational Workshop on Open Data pages
    Leoni
  • On the tradeoff between privacy and utility in data publishing InInternational Conference on Knowledge Discovery and Data Mining pages
    Li
  • a system for evaluating and comparing relational and transaction anonymization algorithms InInternational Conference on Extending Database Technology pages
    Poulis
  • Data privacy through optimal anonymization InInternational Conference on Data pages
    Bayardo
  • Data privacy through optimal anonymization InInternational Conference on Data pages
    Bayardo
  • Comas Enhancing data utility in differential privacy via microaggregation - based anonymity
    Soria
  • Comas Enhancing data utility in differential privacy via microaggregation - based anonymity
    Soria
  • On syntactic anonymity and differential privacy InInternational Conference on Data Engineering Workshops pages
    Clifton
  • Towards a differentially private data anonymization Report
    Fouad
  • On syntactic anonymity and differential privacy InInternational Conference on Data Engineering Workshops pages
    Clifton
  • Distributed data mining with differential privacy International Conference on Communications pages
    Zhang
  • Principles - versus rules - based output statistical disclosure control in remote access environments
    Ritchie
  • processing and machine learning with differential privacy and challenges for continuous data
    Sarwate
  • Empirical privacy and empirical utility of anonymized data InIEEE International Conference on Data Engineering Workshops pages
    Cormode
  • Programs for Publishers
    Quinlan
  • Programs for Publishers
    Quinlan
  • Statistical disclosure control in practice Media
    Willenborg