KAMINO: Constraint-aware differentially private data synthesis

Chang Ge, Shubhankar Mohapatra, Xi He, Ihab F. Ilyas

Research output: Contribution to journalConference articlepeer-review

15 Scopus citations

Abstract

Organizations are increasingly relying on data to support decisions. When data contains private and sensitive information, the data owner often desires to publish a synthetic database instance that is similarly useful as the true data, while ensuring the privacy of individual data records. Existing differentially private data synthesis methods aim to generate useful data based on applications, but they fail in keeping one of the most fundamental data properties of the structured data — the underlying correlations and dependencies among tuples and attributes (i.e., the structure of the data). This structure is often expressed as integrity and schema constraints, or with a probabilistic generative process. As a result, the synthesized data is not useful for any downstream tasks that require this structure to be preserved. This work presents Kamino, a data synthesis system to ensure differential privacy and to preserve the structure and correlations present in the original dataset. Kamino takes as input of a database instance, along with its schema (including integrity constraints), and produces a synthetic database instance with differential privacy and structure preservation guarantees. We empirically show that while preserving the structure of the data, Kamino achieves comparable and even better usefulness in applications of training classification models and answering marginal queries than the state-of-the-art methods of differentially private data synthesis.

Original languageEnglish (US)
Pages (from-to)1886-1899
Number of pages14
JournalProceedings of the VLDB Endowment
Volume14
Issue number10
DOIs
StatePublished - 2021
Externally publishedYes
Event47th International Conference on Very Large Data Bases, VLDB 2021 - Virtual, Online
Duration: Aug 16 2021Aug 20 2021

Bibliographical note

Publisher Copyright:
© by the owner/author(s).

Fingerprint

Dive into the research topics of 'KAMINO: Constraint-aware differentially private data synthesis'. Together they form a unique fingerprint.

Cite this