GermaParl Corpus of Plenary Protocols

The GermaParl Corpus of Parliamentary Protocols has been prepared in the PolMine Project and covers debates in the German Bundestags since the first meeting on September 7, 1949. GermaParl v2.3.0-rc1 covers debates until March 18, 2025. It prepares a forthcoming public release.

GermaParl is a quality-tested resource on German parliamentarism. Newly added material for recent partliamentary sessions has not yet been tested as comprehensively as the data in previous public releases. This is why we call on beta users to be aware of remaining errors in the data and to contribute to the quality of the resource by giving feedback.

Beta users of GermaParl v2.3.0-rc1 need to request access to the corpus and are kindly asked to give feedback on any issues they encounter, so that GermaParl v2.3.0 will be a trustworthy, high-quality resource for research on parliamentary proceedings.

The beta release includes a linguistically annotated indexed version (Corpus Workbench / CWB data format) of GermaParl.

Beta users are requested to proceed as follows:

Request access: Click the respective button on this page. A personal invitation to serve as a beta user is not required to be eligible. A short and telling note on the research interest you associate with GermaParl Beta will help us to make a quick decision.

Confirm Email: You will then receive an email from Zenodo to verify your email-address. It may take a while (up to an hour) until you receive this message. If you still do not find it in your Inbox, check the SPAM folder of your mail account. Confirm your Email address.

Confirmation of access: We need to confirm data access manually. We will consider incoming requests on a continuous basis, but please allow 2-3 days for a response.

Join GitHub issue tracker: To process feedback systematically, we use the issue tracker of a private GitHub repository. We will invite you to be a collaborator of this repository. To be able to invite you to the GitHub repository, we will ask you to provide us with your GitHub account. Please consider creating a GitHub account if you do not yet have one.

Download and install corpus: Once we have confirmed data access, Zenodo will send you an Email with a link. After following the link (“Access the record”), the Zenodo landing page will show multiple downloadable files for the XML and CWB variants of GermaParl under the “Files” heading. If you work with the CWB variant of GermaParl, download the file called “germaparl_v2.3.0-rc1.tar.gz”. A proper internet connection is advisable: The size of the corpus tarball is ~2,6 GB. After the download has finished, identify the path to the downloaded “tar.gz” file. The full path including the file name is used as the “PATH-TO-TARBALL” in the following lines of code. Insert your file