Home > Research > Publications & Outputs > Using saturated models for data synthesis

Research

Associated organisational unit

Electronic data

  • Jackson et al. (2022) IWSM

    Final published version, 711 KB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Keywords

View graph of relations

Using saturated models for data synthesis

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper

Published
Publication date18/07/2022
Host publicationProceedings of the 36th International Workshop on Statistical Modelling: July 18-22, 2022 - Trieste, Italy
EditorsNicola Torelli, Ruggero Bellio, Vito Muggeo
PublisherEUT Edizioni Università di Trieste, Trieste 2022
Pages205-210
Number of pages6
ISBN (Electronic)9788855113090
<mark>Original language</mark>English
Event36th International Workshop on Statistical Modelling: July 18-22, 2022 - Trieste, Italy - Università di Trieste, Trieste, Italy
Duration: 18/07/2022 → 22/07/2022
Conference number: 36
https://www.iwsm2022.com/

Conference

Conference36th International Workshop on Statistical Modelling
Abbreviated titleIWSM
Country/TerritoryItaly
CityTrieste
Period18/07/22 → 22/07/22
Internet address

Conference

Conference36th International Workshop on Statistical Modelling
Abbreviated titleIWSM
Country/TerritoryItaly
CityTrieste
Period18/07/22 → 22/07/22
Internet address

Abstract

The use of synthetic data sets are becoming ever more prevalent,
as regulations such as the General Data Protection Regulation (GDPR), which place greater demands on the protection of individuals’ personal data, are coupled with the conflicting demand to make more data available to researchers. This paper discusses the approach of synthesizing categorical data at the aggregated
(contingency table) level using a saturated count model, which adds noise - and hence protection - to cell counts. The paper also discusses how distributional properties of synthesis models are intrinsic to generating synthetic data with suitable risk and utility profiles.