This project contains the code for the extended abstract Zeitliche Entwicklung parasozialer Beziehungen im Hinblick auf die Dimensionen Valenz und Intensität.

It is divded into 4 main folders:

1. scrape: retrieve all video IDs for a given YouTube channel. This is done by scraping the landing page of the respective channel.
2. data_collection: collect comments and metadata about videos, statistics about the channels.
3. filter_data: conduct data cleaning and preprocessing.
4. analysis: apply LIWC, create and plot models and validate LIWC for positive affect