A de-identification technique alters or limits data so individuals are less readily identifiable. It supports privacy-first data sourcing by reducing re-identification risk while preserving useful information for business purposes. section 2.3.1
Common approaches include:
- Pseudonymization replaces direct identity signals with substitute identifiers. section 2.3.1
- Differential privacy limits what can be learned about individuals from data. section 2.3.1
- K-anonymity reduces re-identification risk when datasets are combined. section 2.3.1
Minimal purchase data
An e-commerce brand stores only the purchase information needed for processing and customer service, omitting broader demographic details without an immediate purpose. section 2.3.1
