Privacy, Data and the Individual. Diferentially Data sets : formal vs empirical approaches to data anonymity

The focus of data anonymity research by computer scientists is almost completely on methods with formal guarantees of anonymity, especially differential privacy. The usefulness of mechanisms with formal guarantees, however, has so far been disappointing. This article argues that computer scientists...

Descripción completa

Detalles Bibliográficos
Autor: Francis, Paul
Tipo de recurso: informe técnico
Fecha de publicación:2019
País:España
Institución:IE
Repositorio:Repositorio IE
OAI Identifier:oai:repositorio.ie.edu:20.500.14417/2767
Acceso en línea:https://doi.org/10.5281/zenodo.3731250
https://hdl.handle.net/20.500.14417/2767
Access Level:acceso abierto
Palabra clave:Data set
Conjuntos de datos
Analysis
Personal Data
Datos personales
Privacy
Privacidad
Marketing
Technology
Tecnología
General Data Protection Regulation
GDPR
Regulación General de Protección de Datos
RGPD
Descripción
Sumario:The focus of data anonymity research by computer scientists is almost completely on methods with formal guarantees of anonymity, especially differential privacy. The usefulness of mechanisms with formal guarantees, however, has so far been disappointing. This article argues that computer scientists should be open to and encouraged to work on empirical data anonymization mechanisms as well—in much the same way that researchers work on both formal and empirical approaches to crypto. This article describes differential privacy and explains its benefits and shortcomings. It also describes a recently developed empirical data anonymization mechanism called Diffix, and describes how transparency and programs that incentivize white-hat attacks, such as bounty programs, can build understanding and confidence in empirical approaches. The article concludes that there is a need for both formal and empirical research on data anonymity.