← Back to glossary
📊

t‑Closeness

Anonymization

Data Ethics Anonymization k‑Anonymity l‑Diversity

Privacy criterion limiting distance between sensitive-attribute distributions in groups vs. overall distribution.

Definition

t‑closeness requires the distribution of a sensitive attribute within any equivalence class to be within a threshold t of the overall distribution (using a distance metric). It aims to reduce attribute disclosure.

In plain English Privacy criterion limiting distance between sensitive-attribute distributions in groups vs. overall distribution.

Why this matters

Why it matters: It addresses weaknesses of k‑anonymity and l‑diversity under skewed data.

Example

Example: Adjust grouping/generalization until each group’s sensitive distribution is close to the global distribution.