Interactive publication
Back to the cataloguePrivacyCase 04 of 12
Differential privacy noise calibration
Adding calibrated noise (e.g. Laplace/Gaussian mechanism) to protect individual privacy in aggregate counts can distort small-cell counts enough to invert rank ordering between nearby cells if the privacy budget (epsilon) is set too aggressively for the use case.
Class
Case to account for
Affects
2 conversions
01
Detection
Compare noised counts against raw (internal-only) counts for rank-order stability across repeated noise draws at the chosen epsilon.
02
Mitigation
01
Tune epsilon per use case, tighter for public releases, looser for internal-only decision support with other controls
02
Report a confidence interval alongside any noised count so users don't over-read small differences
03
Affected conversions
PointH3
OPENRasterH3
OPENMore in Privacy
Consent-based precision reductionDeclared unit below the precise-location thresholdDevice trajectory exposureHousehold-level targeting riskK-anonymity thresholdsMinimum aggregation windowRegional privacy restrictionsResolution degradation policySmall-cell re-identification riskSparse-audience suppressionTemporal leakage