k-Anonymity
It is commonly achieved through generalization or suppression of quasi-identifiers such as age, region, and job title. A dataset with k=5 ensures every combination of those attributes appears in at least five records. k-Anonymity can reduce singling-out risk, but it does not prevent all disclosure, especially when sensitive values are uniform within a group or external information is available.
What is k-Anonymity?
k-Anonymity is a privacy property in which each record is indistinguishable from at least k−1 other records based on selected identifying attributes.
Why it matters
It is commonly achieved through generalization or suppression of quasi-identifiers such as age, region, and job title. A dataset with k=5 ensures every combination of those attributes appears in at least five records. k-Anonymity can reduce singling-out risk, but it does not prevent all disclosure, especially when sensitive values are uniform within a group or external information is available.
Example
Exact ages and postcodes are generalized until every resulting demographic combination appears in at least ten records.