README.md

---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: label
    dtype: string
  - name: num_pages
    dtype: float64
  - name: split
    dtype: string
  splits:
  - name: train
    num_bytes: 254697480
    num_examples: 20818
  download_size: 134126358
  dataset_size: 254697480
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: eupl-1.1
language:
- nl
tags:
- document_classification
size_categories:
- 10K<n<100K
---

This dataset is a collection of Dutch Raadsinformatie documents from the Municipality of Amsterdam, which were published in compliance with the Open Government Act (Woo).
It is used for the Document Classification using Large Language Models project (todo: insert link).
The documents belong to 11 categories, such as Budgets (begrotingen), Research Reports (onderzoeksrapporten) and Motions (moties). 
The project's objective was to classify these documents to make them easier to find and thus improve accessibility.

Modified Version(s) of the Dataset:
- Conversation Dataset to fine-tune LLMs - first 200 tokens per document: [AmsterdamBalancedFirst200Tokens](https://huggingface.co/datasets/FemkeBakker/AmsterdamBalancedFirst200Tokens)

Data sources:
- https://amsterdam.raadsinformatie.nl/
- https://openresearch.amsterdam/
- https://open.amsterdam/

This dataset is part of [inseart thesis info] in collaboration with Amsterdam Intelligence for the City of Amsterdam.