Differentially Private Release of Datasets using Gaussian Copula
We propose a generic mechanism to efficiently release differentially private synthetic versions of high-dimensional datasets with high utility. The core technique in our mechanism is the use of copulas, which are functions representing dependencies among random variables with a multivariate distribu...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Labor Dynamics Institute
2020-06-01
|
| Series: | The Journal of Privacy and Confidentiality |
| Subjects: | |
| Online Access: | https://journalprivacyconfidentiality.org/index.php/jpc/article/view/686 |
| _version_ | 1819024959691816960 |
|---|---|
| author | Hassan Jameel Asghar Ming Ding Thierry Rakotoarivelo Sirine Mrabet Dali Kaafar |
| author_facet | Hassan Jameel Asghar Ming Ding Thierry Rakotoarivelo Sirine Mrabet Dali Kaafar |
| author_sort | Hassan Jameel Asghar |
| collection | DOAJ |
| description | We propose a generic mechanism to efficiently release differentially private synthetic versions of high-dimensional datasets with high utility. The core technique in our mechanism is the use of copulas, which are functions representing dependencies among random variables with a multivariate distribution. Specifically, we use the Gaussian copula to define dependencies of attributes in the input dataset, whose rows are modelled as samples from an unknown multivariate distribution, and then sample synthetic records through this copula. Despite the inherently numerical nature of Gaussian correlations we construct a method that is applicable to both numerical and categorical attributes alike. Our mechanism is efficient in that it only takes time proportional to the square of the number of attributes in the dataset. We propose a differentially private way of constructing the Gaussian copula without compromising computational efficiency. Through experiments on three real-world datasets, we show that we can obtain highly accurate answers to the set of all one-way marginal, and two-and three-way positive conjunction queries, with 99% of the query answers having absolute (fractional) error rates between 0.01 to 3%. Furthermore, for a majority of two-way and three-way queries, we outperform independent noise addition through the well-known Laplace mechanism. In terms of computational time we demonstrate that our mechanism can output synthetic datasets in around 6 minutes 47 seconds on average with an input dataset of about 200 binary attributes and more than 32,000 rows, and about 2 hours 30 mins to execute a much larger dataset of about 700 binary attributes and more than 5 million rows. To further demonstrate scalability, we ran the mechanism on larger (artificial) datasets with 1,000 and 2,000 binary attributes (and 5 million rows) obtaining synthetic outputs in approximately 6 and 19 hours, respectively. These are highly feasible times for synthetic datasets, which are one-off releases. |
| first_indexed | 2024-12-21T05:03:04Z |
| format | Article |
| id | doaj.art-6a224679836f448683a4f6bee112b8be |
| institution | Directory Open Access Journal |
| issn | 2575-8527 |
| language | English |
| last_indexed | 2024-12-21T05:03:04Z |
| publishDate | 2020-06-01 |
| publisher | Labor Dynamics Institute |
| record_format | Article |
| series | The Journal of Privacy and Confidentiality |
| spelling | doaj.art-6a224679836f448683a4f6bee112b8be2022-12-21T19:15:12ZengLabor Dynamics InstituteThe Journal of Privacy and Confidentiality2575-85272020-06-0110210.29012/jpc.686Differentially Private Release of Datasets using Gaussian CopulaHassan Jameel Asghar0Ming Ding1Thierry Rakotoarivelo2Sirine Mrabet3Dali Kaafar4Macquarie UniversityData61, CSIROData61, CSIROData61, CSIROMacquarie University and Data61, CSIROWe propose a generic mechanism to efficiently release differentially private synthetic versions of high-dimensional datasets with high utility. The core technique in our mechanism is the use of copulas, which are functions representing dependencies among random variables with a multivariate distribution. Specifically, we use the Gaussian copula to define dependencies of attributes in the input dataset, whose rows are modelled as samples from an unknown multivariate distribution, and then sample synthetic records through this copula. Despite the inherently numerical nature of Gaussian correlations we construct a method that is applicable to both numerical and categorical attributes alike. Our mechanism is efficient in that it only takes time proportional to the square of the number of attributes in the dataset. We propose a differentially private way of constructing the Gaussian copula without compromising computational efficiency. Through experiments on three real-world datasets, we show that we can obtain highly accurate answers to the set of all one-way marginal, and two-and three-way positive conjunction queries, with 99% of the query answers having absolute (fractional) error rates between 0.01 to 3%. Furthermore, for a majority of two-way and three-way queries, we outperform independent noise addition through the well-known Laplace mechanism. In terms of computational time we demonstrate that our mechanism can output synthetic datasets in around 6 minutes 47 seconds on average with an input dataset of about 200 binary attributes and more than 32,000 rows, and about 2 hours 30 mins to execute a much larger dataset of about 700 binary attributes and more than 5 million rows. To further demonstrate scalability, we ran the mechanism on larger (artificial) datasets with 1,000 and 2,000 binary attributes (and 5 million rows) obtaining synthetic outputs in approximately 6 and 19 hours, respectively. These are highly feasible times for synthetic datasets, which are one-off releases.https://journalprivacyconfidentiality.org/index.php/jpc/article/view/686differential privacy; synthetic data; high dimensional; copula |
| spellingShingle | Hassan Jameel Asghar Ming Ding Thierry Rakotoarivelo Sirine Mrabet Dali Kaafar Differentially Private Release of Datasets using Gaussian Copula The Journal of Privacy and Confidentiality differential privacy; synthetic data; high dimensional; copula |
| title | Differentially Private Release of Datasets using Gaussian Copula |
| title_full | Differentially Private Release of Datasets using Gaussian Copula |
| title_fullStr | Differentially Private Release of Datasets using Gaussian Copula |
| title_full_unstemmed | Differentially Private Release of Datasets using Gaussian Copula |
| title_short | Differentially Private Release of Datasets using Gaussian Copula |
| title_sort | differentially private release of datasets using gaussian copula |
| topic | differential privacy; synthetic data; high dimensional; copula |
| url | https://journalprivacyconfidentiality.org/index.php/jpc/article/view/686 |
| work_keys_str_mv | AT hassanjameelasghar differentiallyprivatereleaseofdatasetsusinggaussiancopula AT mingding differentiallyprivatereleaseofdatasetsusinggaussiancopula AT thierryrakotoarivelo differentiallyprivatereleaseofdatasetsusinggaussiancopula AT sirinemrabet differentiallyprivatereleaseofdatasetsusinggaussiancopula AT dalikaafar differentiallyprivatereleaseofdatasetsusinggaussiancopula |