This part has been test run in the client container deployed as described by the handout's README. You may need to install the required package with pip install -r requirements.txt
The privacy evaluation was fully conducted in the privacy_evaluation/PrivacyEval.ipynb notebook. It uses the provided privacy_evaluation/grid.py and privacy_evaluation/queries.csv, privacy_evaluation/pois.csv files.
It is divided in the following parts:
- Package Imports: the necessary package imports for the privacy evaluation.
- Data Imports & Examination: Importing the csv files and examining the contents as well as extracting general statistics.
- Data Analysis: Creating visualisation for the data.
- Data Improvements: enhancing the data (e.g. timestamps and coordinates)
- Data Analysis (Attack): extracting insight from the data to perform attacks and comparing results given certain mitigations.
- Defence: explanation of the mitigations.