DigiNews

Tech Watch Articles

← Back to articles

Re-Identification Risk vs k-Anonymity

Quality: 8/10 Relevance: 9/10

Summary

The article provides an experimental analysis of k-anonymity strength on a synthetic dataset to measure its impact on re-identification risk and data utility. It shows that increasing k reduces attacker success (Hit@1) but causes non-linear utility loss, with a notable sweet spot around moderate k; it also discusses attacker models and recommends combining k-anonymity with other techniques to improve resilience.

🚀 Service construit par Johan Denoyer