Project: Bundestag / German parliament

Final project for the lecture Visualization.

Task

this task is formulated for the German parliament, but can of course be applied/translated to other countries as well

gather data about the members of the Bundestag, including party affiliation, date of birth, Wahlkreis or Landesliste, since when they are members of parliament

each MP should have a small profile page

each party should have a profile page, giving the number of MPs and their list

a global alphabetical list of MPs

other possible task: for some (possibly fictional) votes show how each member voted, for each party give statistics on how their members voted

realization like the bibliography network task for instance as static HTML file collection via XML and Python

Website

You can access a hosted version of the website on https://njbirth.github.io./visualization.

If you want to run it yourself, you need node.js and npm to be installed on your system. After that you can just switch to the website folder, run npm install and npm run dev. After some compilation steps, you should be able to access the website on localhost:5000. If you run into memory issues, try export NODE_OPTIONS=--max_old_space_size=4096.

Data

Sources

The version of the MP data we used is from 2021-03-12. Please note that Nikolas Löbel left the parliament on 2021-03-10, while his replacement, Kordula Kovac, joined on 2021-03-17. Therefore, in our data the Bundestag has only 708 MPs and not 709.

See Acknowledgements.svelte for more information (or https://njbirth.github.io/visualization -> Quellen und Lizenzen, if you want a prettier version of the file).

Directory structure

/data
├── _raw
│   ├── votes
|   |   └── ...
│   └── MDB_STAMMDATEN.XML
├── __converted
│   ├── votes
|   |   ├── meta.json
|   |   └── ...
│   └── MDB_STAMMDATEN.json
├── ___final
│   ├── meta.json
│   ├── votes.json
│   └── MDB_STAMMDATEN.json
├── img
│   └── ...
├── map
│   └── ...
└── preprocess_data.ipynb

The data in the raw folder is directly taken from the above mentioned sources. We converted this files to JSON/CSV, because we found it more convenient to work with (using this XML to JSON converter; for XLSX to CSV we just opened the file with LibreOffice Calc and saved it again as CSV). The final folder contains the data that we got after further processing of the converted data (also see preprocess_data.ipynb). __converted/votes/meta.json contains metadata to the votes, which is used by the Jupyter notebook. ___final/meta.json was generated by hand and contains some party metadata for the website. (The underscores in the directory names are just for the alphabetical ordering). img contains URLs for the MP portraits and a notebook to obtain then. map includes election data that is used for the map (kind of obvious).

oleuml / visualization