Protecting micro-data by micro-aggregation: the experience in Eurostat

A natural strategy to protect the confidentiality of individual data is to aggregate them at the lowest possible level. Some studies realised in Eurostat on this topic will be presented: properties of classifications in clusters of fixed sizes, micro-aggregation as a generic method to protect the co...

ver descrição completa

Detalhes bibliográficos
Autor: Defays, Daniel
Tipo de documento: artigo
Data de publicação:1997
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositório:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglês
OAI Identifier:oai:upcommons.upc.edu:2099/4685
Acesso em linha:https://hdl.handle.net/2099/4685
Access Level:Acceso aberto
Palavra-chave:Statistics
Confidentiality
Clustering
Micro-aggregation
Aplicacions (Matemàtica)
Classificació AMS::62 Statistics::62P Applications
Descrição
Resumo:A natural strategy to protect the confidentiality of individual data is to aggregate them at the lowest possible level. Some studies realised in Eurostat on this topic will be presented: properties of classifications in clusters of fixed sizes, micro-aggregation as a generic method to protect the confidentiality of individual data, application to the Community Innovation Survey. The work performed in Eurostat will be put in line with other projects conducted at European level on the topic of statistical confidentiality.