Data Privacy in Practice: Pseudonymisation, k-Anonymity & Re-identification Risk
Prepare a (fictional, synthetic) health-survey extract for safe sharing: identifier types, pseudonyms and salted IDs, generalisation, k-anonymity with COUNTIFS, suppression and information loss, l-diversity, re-identification risk and a retention schedule, every cell checked live. Educational, not legal advice.
The first lesson is free — no account needed.
About this course
A hands-on data privacy course that prepares a fictional, synthetic health-survey extract for safe sharing, for analysts, researchers and privacy officers who need to see the numbers behind anonymisation. You sort columns into direct identifiers, quasi-identifiers and sensitive attributes, replace names with pseudonyms from a key table, build an illustrative salted ID and mask emails, coarsen ages into bands and ZIP codes into prefixes, measure equivalence classes and the k of a release with COUNTIFS, suppress small classes and price the information loss, test l-diversity, turn class sizes into average and maximum re-identification risk and apply a retention schedule with EDATE. Laws such as the GDPR are referred to only in general terms. Educational, not legal advice. The first lab is free.
Lessons
- Direct Identifiers, Quasi-Identifiers & Sensitive AttributesFree
- Pseudonymisation: Lookup Keys, Salted IDs & Masking Pro
- Generalisation: Age Bands & ZIP Prefixes Pro
- k-Anonymity: Equivalence Classes with COUNTIFS Pro
- Suppression of Small Classes & Information Loss Pro
- l-Diversity: When Everyone in a Class Shares the Secret Pro
- Re-identification Risk: 1/k, Average & Maximum Risk Pro
- Data Minimisation & Retention: Which Rows Must Go? Pro