Contains Code of Privacy Computation Demo presented at CCS'18 conference.
The privacy web-page is built on top of the python web framework django and can be started from the directory "webpage" with the command "python3 manage.py runserver"
If you have intentions with this code beyond playig around, please do not forget to change the Django Secret Key!
For addidional information on Probability Buckets, see the Github Repo, and realted publications [1] and [2].
[1] S. Meiser, E. Mohammadi, "Tight on Budget? Tight Bounds for r-Fold Approximate Differential Privacy", Proceedings of the 25th ACM Conference on Computer and Communications Security (CCS), 2018 (https://eprint.iacr.org/2017/1034.pdf)
[2] Sommer DM, Meiser S, Mohammadi E. Privacy loss classes: The central limit theorem in differential privacy. Proceedings on Privacy Enhancing Technologies. 2019 Apr (https://eprint.iacr.org/2018/820.pdf)