1. Import the original data
The original dataset must be in a simple text file with any type of delimiter, i.e., a character separating different fields for each field of a record. Amnesia has an import wizard, which guesses the type of the data and asks the user to confirm it. The user selects which fields that will participate in the process and which will be left out. You can find examples of datasets here.
2. Create the generalization hierarchies
The basic idea behind the anonymization of a dataset is to replace unique values or unique combinations of values, e.g., zip code and date of birth, with more abstract ones, e.g., city and year of birth, so that they are no longer identified. Amnesia allows a user to create these rules for generalizing values in a semi-automatic way, to save them and re-use them or import them from other sources.
3. Ready? Go
The user can select the most suitable method for this problem (e.g., k-anonymity or km-anonymity and link the hierarchies with the respective attributes of the records. The anonymization process starts.
4. Choose the solution you like
Amnesia depicts several possible solutions, it visualizes the distribution of values and it provides statistics about the data quality in the anonymous dataset. By simple clicks a solution tailored to user needs can be inspected and applied.
5. You are ready!
The anonymized data can be saved locally or directly to Zenodo!