Data anonymization is a method of protecting personal data by modifying or removing any information that can be used to identify an individual, either directly or indirectly, in a data set. Data anonymization aims to prevent the re-identification of the data subjects, even by the data controller or processor, or by using additional data sources or techniques. Data anonymization also helps to comply with data protection laws and regulations, such as the General Data Protection Regulation (GDPR) or the California Consumer Privacy Act (CCPA), which require data controllers and processors to respect the privacy rights and preferences of the data subjects.
The data is transformed such that re-identification is impossible is an example of data anonymization, as it involves applying irreversible techniques, such as aggregation, generalization, perturbation, or synthesis, to alter the original data in a way that preserves their utility and meaning, but eliminates their identifiability. For example, a database of customer transactions can be anonymized by replacing the names and addresses of the customers with random codes, and by adding noise or rounding to the amounts and dates of the transactions.
The other options are not examples of data anonymization, but of other methods of protecting personal data that do not guarantee the impossibility of re-identification. The data is encrypted and a key is required to re-identify the data is an example of data pseudonymization, which is a method of replacing direct identifiers with pseudonyms, such as codes or tokens, that can be linked back to the original data with a key or algorithm. Data pseudonymization does not prevent re-identification by authorized parties who have access to the key or algorithm, or by unauthorized parties who can break or bypass the encryption. Key fields are hidden and unmasking is required to access to the data is an example of data masking, which is a method of concealing or obscuring sensitive data elements, such as names or credit card numbers, with characters, symbols or blanks. Data masking does not prevent re-identification by authorized parties who have permission to unmask the data, or by unauthorized parties who can infer or guess the hidden data from other sources or clues. Names and addresses are removed but the rest of the data is left untouched is an example of data deletion, which is a method of removing direct identifiers from a data set. Data deletion does not prevent re-identification by using indirect identifiers, such as age, gender, occupation or location, that can be combined or matched with other data sources to re-establish the identity of the data subjects.
[References:, Big Data Deidentification, Reidentification and Anonymization - ISACA, section 2: “Anonymization is the ability for the data controller to anonymize the data in a way that it is impossible for anyone to establish the identity of the data.”, Data Anonymization - Overview, Techniques, Advantages, section 1: “Data anonymization is a method of ensuring that the company understands and enforces its duty to secure sensitive, personal, and confidential data in a world of highly complex data protection mandates that can vary depending on where the business and the customers are based.”, ]
