k-anonymity check on a set of quasi-identifiers

A release satisfies k-anonymity when every combination of quasi-identifier values that appears in it appears at least k times. The equivalence classes are the distinct rows of the quasi-identifier block; the smallest is the binding constraint, and the rows in classes below k are exactly the records that would have to be suppressed or generalised.

Usage

Kanon(y, quasi_ids, k)

Arguments

y

Records; only the length is used, to check it against the quasi-identifier block.

quasi_ids

Quasi-identifier block, one row per record.

k

Required anonymity level.

Value

List with estimate (smallest class size), min_class_size, max_class_size, mean_class_size, k, satisfies, n_classes, n_violating, n_unique, n, method.

Details

Formula: min over equivalence classes of |class| >= k.

References

Sweeney (2002), k-anonymity: a model for protecting privacy, International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems 10(5):557-570. doi:10.1142/S0218488502001648

Examples

Kanon(y = c(1, 2, 3, 4, 5, 6, 7, 8), quasi_ids = c(1, 2, 3, 4, 5, 6, 7, 8), k = 5L)
#> $estimate
#> [1] 1
#> 
#> $min_class_size
#> [1] 1
#> 
#> $max_class_size
#> [1] 1
#> 
#> $mean_class_size
#> [1] 1
#> 
#> $k
#> [1] 5
#> 
#> $satisfies
#> [1] 0
#> 
#> $n_classes
#> [1] 8
#> 
#> $n_violating
#> [1] 8
#> 
#> $n_unique
#> [1] 8
#> 
#> $n
#> [1] 8
#> 
#> $method
#> [1] "min over equivalence classes of |class| >= k, Sweeney (2002)"
#> 
#> attr(,"class")
#> [1] "morie_rich_result" "list"