Generate a privacy report comparing real and synthetic data

Generate a privacy report comparing real and synthetic data

Usage

privacy_report(real, synthetic, sensitive_col = NULL, known_cols = NULL)

Arguments

real

A data frame of real data.

synthetic

A data frame of synthetic data.

sensitive_col

Optional. Column name for attribute disclosure risk.

known_cols

Optional. Column names known to an adversary (required if sensitive_col is supplied).

Value

An rsdv_privacy_report object.

Examples

# \donttest{
syn   <- gaussian_copula_synthesizer(metadata(adult_income)) |>
  fit(adult_income)
synth <- sample(syn, n = 500)
pr    <- privacy_report(adult_income, synth)
print(pr)
#> == rsdv Privacy Report ==
#> 
#> NNDR Score (higher = more private):  0.992
# }