Abstract
This paper gives a systematic overview of the methods available for protecting sensitive data, organised around the different kinds of analysis the data may need to support. It spans the main approaches — statistical disclosure control, anonymisation, synthetic data, and secure computation and access models — and discusses how the choice depends on the intended use and the threat model. By mapping methods to use cases, it helps data holders reason about which protection strategy fits a given release. Published in the International Journal of Information Security, it connects the statistical and computer-science traditions of data protection.
anonymization privacy sensitive data statistical disclosure control