German Parliamentary Corpus (GerParCor)

This corpus contains (mostly historical) German-language parliamentary proceedings from the 19th, 20th, and 21th centuries, including state and federal-level data. Additionally, the corpus contains conversions of scanned protocols and, in particular, of protocols in [Fraktur](https://en.wikipedia.org/wiki/Fraktur) converted via an OCR process based on [Tesseract](https://github.com/tesseract-ocr/tesseract). All protocols were preprocessed by means of the NLP pipeline [spaCy v3](https://spacy.io/usage/v3/) and automatically annotated with metadata regarding their session date. The corpus is made available in the XML format of the [UIMA project](https://uima.apache.org/).
The corpus is available for download from GitHub.