What is the process of removing or modifying personal information from a dataset called?

The process of removing or modifying personal information from a dataset is referred to as de-identification. This term encompasses various techniques that aim to prevent the identification of individuals while still allowing the data to be utilized for analysis.

De-identification typically involves either anonymization, where all personally identifiable information is stripped from the dataset, or pseudonymization, where identifiable information is replaced with pseudonyms or codes. The core goal is to protect individual privacy while facilitating data use for research, policy analysis, or other purposes.

In contrast, data minimization is a principle that emphasizes collecting only the personal data that is necessary for a specific purpose, rather than focusing specifically on the alteration of data already collected. Anonymization is often seen as a subset of de-identification, specifically referring to the complete removal of identifiable features. Decentralization, meanwhile, pertains to the distribution of data or control rather than the modification or removal of information. Understanding these distinctions is crucial for applying privacy practices appropriately and effectively.