Design of synthetic data generation methods preserving
certain relationships between the original data for a number
of subdomains (subsets of variables and/or records).
of the synthetic data generators to the use of synthetic
log files in data warehouses.
Design of re-identification
methods (via record linkage) to assess the level of privacy
provided by the synthetic
Design of utility measures for the synthetic
data with respect to the original data. This objective
will be pursued
in co-operation with CRISES/URV, who are experienced
in the utility analysis of perturbed data.
Dr. Vicenc Torra (CT, IIIA-CSIC), principal researcher
(IP) of the subproject