Quasiidentifiable

Quasiidentifiable is an adjective used in data privacy to describe attributes or data sets that do not identify a person on their own but can become identifying when combined with other information. In privacy research, quasi-identifiers (or quasi-identifiable data) are attributes that, taken together with external data sources, can significantly narrow down or reveal an individual’s identity.

Examples commonly cited as quasi-identifiers include date of birth, gender, ZIP or postal code, and occupation. Each attribute may not uniquely identify a person, but their combination can reduce the pool of possible identities to a single individual in many contexts. The identifiability of quasi-identifiers is highly context-dependent, varying with the availability of external data and the adversary’s linking capabilities.

Re-identification risk is the central concern with quasi-identifiables. Even datasets that are de-identified or anonymized can be vulnerable if quasi-identifiers are present and external information is accessible. This risk has driven the development of privacy techniques such as k-anonymity, l-diversity, t-closeness, differential privacy, and other data minimization and transformation approaches.

Mitigation strategies focus on reducing precision, restricting access, or adding uncertainty. Examples include generalizing specific values (turning exact birth dates into birth years), suppressing some attributes, adding noise, providing controlled data environments, and applying formal privacy guarantees.

Context and governance: Depending on jurisdiction, quasi-identifiers may be treated as protecting personal data or as PII-like information, subject to data protection rules and ethical considerations. The concept highlights that identifiability is not absolute and depends on data context and external sources.