The DutchParl dataset of parliamentary proceedings is available for research. It can be downloaded at this link with password DutchParl2026. The dataset contains data from several countries, all in the same XML format. These are in the folder permanent-> archive. Then use the country codes to select the collection.
Dutch Parliamentary proceedings
If you want the Dutch proceedings, go to the nl folder, and then choose the three tar files starting with d-nl-proc. The d-nl-proc-sgd file contains the proceedings from before 1995 and which are scans with not always very good OCR.
You may also want to look at the m-nl and p-nl files which contain metadata about the politicians and parties occuring in the dataset. The speeches in the XML files are labeled by their speakers and their party affiliations and the identifiers refer back to these metadata files
Parlamint
The DutchParl dataset was a predecessor of the Parlamint dataset, which is much larger, better and richer annotated. So, if you want to use parliamentary data, first go there. There are some differences in temporal scope, though. The DutchParl for the Netherlands contains much older proceedings, and this may also hold for the other countries.
Citation
If you use the DutchParl dataset, it would be great if you can cite our paper on it. Here is its Google Scholar Page with other papers using this dataset. Thanks!
Marx, Maarten, and Anne Schuth. “DutchParl. A corpus of parliamentary documents in Dutch.” Proceedings Language Resources and Evaluation (LREC) pp (2010): 3670-3677.
