Data Privacy Concerns - A Risk Based Approach to De-Identification
summary
Comcast builds a de-identification center of excellence, implements a 10-step process and a dashboarded risk engine that scores public and acquaintance re‑identification risk, uses the higher of those risks plus context and a 0.04% uniqueness threshold to declare anonymization, and runs the full quasi‑identifier set (not a sample) through automated transformations until the numerical threshold is met. Results: a repeatable certification, internal service to transform vendor and partner datasets, longer retention and sharing for analytics, CI/CD integration, and no approved exceptions so far.
AI-written summary · about
chapters (4)
- 0:00Start
- 4:26Some Privacy Functional Areas
- 7:13Privacy Organization Components
- 14:18De-Identification Risk Components
related talks
09mPrivacy Analytics and De-Identification
32mThe Dangerous Data Anonymization - Keynote
10mHealth Data Anonymization and Privacy
29mPersonal Data Pseudonymization Versus Anonymization In The Age Of AI & Big Data
01h 17mTo be or not to be (anonymous)? Anonymyity in the age of big and open data.
16mBuilding an End-to-End De-Identification Pipeline for Advertising Activity Data at...
15mDe-identified, Anonymized, or Aggregated? A Look around the Globe
01h 13mThe risk-based approach to privacy: How far should we go?