Privkit: A Privacy Toolkit

Privkit is a privacy toolkit that provides methods for privacy analysis. It includes different data types, privacy-preserving mechanisms, attacks, and metrics. The current version is focused on location data and facial data. The Python Package is designed in a modular manner and can be easily extended to include new mechanisms. Privkit can be used to process data, configure privacy-preserving mechanisms, apply attacks, and also evaluate the privacy/utility trade-off through suitable metrics.

Warning

privkit is an ongoing open-source framework. If you find errors or if you have suggestions, please open an issue in the repository. We would love to hear from you!

• User Guide
• Github
• Tutorials
• Web Version

Citation

If you use Privkit in your work, please consider to cite:

Cunha, M., Duarte, G., Andrade, R., Mendes, R., & Vilela, J. P. (2024, June). Privkit: A Toolkit of Privacy-Preserving Mechanisms for Heterogeneous Data Types. In Proceedings of the Fourteenth ACM Conference on Data and Application Security and Privacy (pp. 319-324).

@inproceedings{cunha2024privkit,
  title={Privkit: A Toolkit of Privacy-Preserving Mechanisms for Heterogeneous Data Types},
  author={Cunha, Mariana and Duarte, Guilherme and Andrade, Ricardo and Mendes, Ricardo and Vilela, Jo{\~a}o P},
  booktitle={Proceedings of the Fourteenth ACM Conference on Data and Application Security and Privacy},
  pages={319--324},
  year={2024}
}

Acknowledgments

We would like to acknowledge the support of PRIVATEER.