Anonymisation-Techniques

This code was written for the Anonymity Techniques assignment of the course Privacy Enhancing Technologies given at the University of Twente.

To view the assignment, see 03_anonymisation.pdf.

Running the Code

• Go to [https://nodejs.org/en/download/package-manager] and follow the installation instructions for your specific operating system.
• Clone this repository to a location of your choosing.
• Open your terminal (if on Windows: command prompt).
• In your terminal, navigate to the folder in which you cloned this repo. (In other words, where package.json is located).
• Enter the following command: npm start.
• Follow the instructions in your terminal.

Files

• package.json: Contains information NodeJS needs to run the application.
• node_modules/: Contains libraries used by application.
• src/Assignments.js: In this file the actual anonymisation techniques are used to anonymize the dataset and perform the tasks specified in the assignment.
• src/CLI.js: Manages the command line interface used to interact with the program, no assignment specific stuff happens here.
• src/Dataloader.js: Reads the data file and parses the values within it. Entries with unknown values are automatically skipped.
• src/Generalization.js: Provides the functionality to perform generalization on the adult dataset.
• src/LocalSuppression.js: Provides the functionality to perform suppression on the adult dataset.
• src/Precision.js: Is used to calculate the precision of generalized dataset.
• src/Stats.js: Provides the functionality to calculate various statistics of a dataset for each DGH. Statistics such as number of categories a DGH has in a dataset, mean size of that category, size of smallest category and the categories themselves along with number of entries in each category.
• data/adult.data: The actual dataset.
• data/cols.json: A simple mapping of the attribute names (age, race, etc.) to their indices in the dataset.
• data/DGHs.json: The DGHs as specified in section \ref{sec:domain_generalization_hierarchies.
• data/public_data.json: The public data used for assignment 1 and 5.
• data/QIDs.json: The attribute names of the quasi identifiers.
• data/SEs.json: The attribute names of the sensitive attributes.