By anonymizing data, we protect the identity of an individual from disclosing sensitive personal information. Hospitals store data on their patients in databases. Many of stored data is used for research purposes. According to the laws, personal data of patients must be anonymised in order to ensure the patient’s privacy. In this thesis we compared a set of anonymization methods on the synthetically generated database. The database was composed of Slovenian patients, where sensitive personal information of each patient is their cholesterol level. We used the ARX software tool to perform anonymization methods. When comparing the methods, we measured the speed of the algorithm and the quality of anonymous data.
|
