consulting: An HTML repository from lisa-wm

Topic-specific sentiment analysis for tweets by German MPs

This repository is structured as follows:

1_scraping contains all Jupyter scripts for scraping the data.
- 1_input contains a manually created list of Twitter accounts owned by AfD members as well as the chromedriver application needed for scraping. The one stored works for Windows OS and a particular Chrome version; please make sure to keep the appropriate one there.
- All scripts that must be executed to obtain the data are in 2_scripts_python and prefixed by the order of execution.
- Results are stored in 3_output.
2_code contains all R and Python files for the actual analysis. A separate readme file there gives details on how to run the code.
3_presentation contains the presentation and associated files.
4_report contains the report and associated files.
5_seminar contains all files created for the teaching material and seminar. The website, however, is linked to another repo (https://github.com/lisa-wm/nlp-twitter-r-bert).