Abstract: A method, system and computer memory for optimally de-identifying a dataset is provided. The dataset from a storage device. The equivalence classes within the dataset is determined. A lattice is determined defining anonymization strategies. A solution set for the lattice is generated. Optimal node from the solution set is determined. The dataset is then de-identified using the generalization defined by the optimal node and can then be stored on the storage device.

Abstract: A system and method of performing risk assessment of a dataset de-identified from a source database containing information identifiable to individuals is provided. The de-identified dataset is retrieved comprising a plurality of records from a storage device. A selection of variables from a user is received, the selection made from a plurality of variables present in the dataset, wherein the variables are potential identifiers of personal information. A selection of a risk threshold acceptable for the dataset from a user is received. A selection of a sampling fraction wherein the sampling fraction define a relative size of their dataset to an entire population is received. A number of records from the plurality of records for each equivalence class in the identification dataset for each of the selected variables. A re-identification risk using the selected sampling fraction is calculated. The re-identification risk meets the selected risk threshold is determined.

Abstract: Disclosures of databases for secondary purposes is increasing rapidly and any identification of personal data may from a dataset of database can be detrimental. A re-identification risk metric is determined for the scenario where an intruder wishes to re-identify as many records as possible in a disclosed database, known as a marketer risk. The dataset can be analyzed to determine equivalence classes for variables in the dataset and one or more equivalence class sizes. The re-identification risk metric associated with the dataset can be determined using a modified log-linear model by measuring a goodness of fit measure generalized for each of the one or more equivalence class sizes.

Abstract: A method, system and computer memory for optimally de-identifying a dataset is provided. The dataset from a storage device. The equivalence classes within the dataset is determined. A lattice is determined defining anonymization strategies. A solution set for the lattice is generated. Optimal node from the solution set is determined. The dataset is then de-identified using the generalization defined by the optimal node and can then be stored on the storage device.

Abstract: A system and method of performing risk assessment of a dataset de-identified from a source database containing information identifiable to individuals is provided. The de-identified dataset is retrieved comprising a plurality of records from a storage device. A selection of variables from a user is received, the selection made from a plurality of variables present in the dataset, wherein the variables are potential identifiers of personal information. A selection of a risk threshold acceptable for the dataset from a user is received. A selection of a sampling fraction wherein the sampling fraction define a relative size of their dataset to an entire population is received. A number of records from the plurality of records for each equivalence class in the identification dataset for each of the selected variables. A re-identification risk using the selected sampling fraction is calculated. The re-identification risk meets the selected risk threshold is determined.