This code was written as part of the fulfilment of the master's thesis of the Data Science & Society programme at Tilburg University.

Recreating the environment

To recreate the environment, paste the following command in the terminal:

conda create --name <env> --file requirements.txt

NOTE: Not all code was run on the same machine, e.g. some code was run in an Ubuntu server environment, some code was run locally on MacOS. It is therefore possible that not all of the code in this repository will run as is.

Running the code

For an overview of the order in which scripts must be run, including some additional code that was not run within these scripts, please refer to main.ipynb.

Other relevant code

• Feature generation: helpers.py > prepare_for_ml().
• Specific steps during preprecessing: please refer to the commented out with # FULL PIPELINE in the first cell of main.ipynb.