Anonymisation by generalisation (k-anonymity)
Measure k before and after generalisation and the utility cost.
العربيةThe idea
Generalise age and postcode, suppress fields, and watch the smallest equivalence class and the share of unique records.
Data description
Synthetic data generated in your browser from stated distributions; no real personal data.
Suggested next experiment
Concepts in this experiment
k-anonymity (إخفاء الهوية k)
Every record matches at least k − 1 others on quasi-identifiers, via generalisation or suppression. It protects against identity linkage but not attribute disclosure when the sensitive value is uniform in a group.
Scientific sources for this experiment
- Simple Demographics Often Identify People UniquelyL. Sweeney (2000) — Paper
