Data de-identification across different data sources using a common data model
View Patent ↗A computer system migrates and de-identifies data. Data is migrated from a dataset to a common data model that is configured to accommodate data comprising a plurality of different data types to be de-identified. Data is analyzed in the common data model to identify privacy vulnerabilities and determine corresponding data de-identification techniques and configuration options to be applied to the data. The automatically determined data de-identification techniques are applied to the data to address all of the identified privacy vulnerabilities, and the resulting de-identified data is migrated from the common data model back to the dataset. Embodiments of the present invention further include a computer-implemented method and program product for migrating and de-identifying data in substantially the same manner described above.
1. A method, in a data processing system comprising at least one processor and at least one memory, the at least one memory comprising instructions executed by the at least one processor to cause the at least one processor to migrate and de-identify data, the method comprising:
migrating data of a dataset to be de-identified to a common data model, wherein the common data model is configured to accommodate data comprising a plurality of different data types to be de-identified, and wherein the common data model comprises a plurality of tables to store corresponding direct identifiers in the dataset, quasi-identifiers in the dataset, sensitive attributes in the dataset, and links between attributes in the plurality of tables and corresponding domain generalization attribute hierarchies;
analyzing the data in the common data model to automatically identify privacy vulnerabilities and determine corresponding data de-identification techniques and configuration options to be applied to the data;
applying the automatically determined data de-identification techniques to the data that resides in the common data model to address all the identified privacy vulnerabilities and produce de-identified data; and
migrating the de-identified data from the common data model back to the dataset.
2. The method of claim 1 , wherein the plurality of tables of the common data model further store relational data, transaction data, sequential data, and location and mobility data.
3. The method of claim 1 , wherein the attribute domain generalization hierarchies specify acceptable generalization operations of corresponding attributes of the data for de-identification.
4. The method of claim 1 , wherein analyzing the data in the common data model further comprises:
processing the direct identifiers and quasi identifiers of the data according to privacy requirements; and
determining a privacy risk of results of the processing to identify an extent of protection of the identified privacy vulnerabilities.
5. The method of claim 3 , wherein analyzing the data in the common data model further comprises generalizing the data using the attribute domain generalization hierarchies.
6. The method of claim 1 , wherein data is automatically migrated from the dataset to the common data model using schema matching techniques.