Thirty years of optimization-based SDC methods for tabular data

In 1966 Bacharach published in Management Science a work on matrix rounding problems in two-way tables of economic statistics, formulated as a network optimization problem. This is likely the first application of optimization/operations research for statistical disclosure control (SDC) in tabular da...

Descripción completa

Detalles Bibliográficos
Autor: Castro Pérez, Jordi|||0000-0003-3573-4568
Tipo de recurso: artículo
Fecha de publicación:2023
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/387618
Acceso en línea:https://hdl.handle.net/2117/387618
Access Level:acceso abierto
Palabra clave:Linear optimization
Mixed integer linear optimization
Network optimization
Statistical disclosure control
Cell suppression
Controlled adjustment
Controlled perturbation
Classificació AMS::90 Operations research, mathematical programming
Àrees temàtiques de la UPC::Matemàtiques i estadística::Investigació operativa
Descripción
Sumario:In 1966 Bacharach published in Management Science a work on matrix rounding problems in two-way tables of economic statistics, formulated as a network optimization problem. This is likely the first application of optimization/operations research for statistical disclosure control (SDC) in tabular data. Years later, in 1982, Cox and Ernst used the same approach in a work in INFOR for a similar problem: controlled rounding. And thirty years ago, in 1992, a paper by Kelly, Golden and Assad appeared in Networks about the solution of the cell suppression problem, also using network optimization. Cell suppression was used for years as the main SDC technique for tabular data, and it was an active field of research which resulted in several lines of work and many publications. The above are some of the seminal works on the use of optimization methods for SDC when releasing tabular data. This paper discusses some of the research done this field since then, with a focus on the approaches that were of practical use. It also discusses their pros and cons compared to recent techniques that are not based on optimization methods.