README.md

---
dataset_info:
  features:
  - name: word
    dtype: string
  - name: pos
    dtype: string
  splits:
  - name: train
    num_bytes: 11682
    num_examples: 704
  download_size: 5014
  dataset_size: 11682
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

## Dataset Description

This dataset contains **example sentences** extracted from a piece of historical Danish newspaper fiction published in **1824**. The text is randomly selected from the [Press-and-Plot dataset](https://huggingface.co/datasets/chcaa/Press-and-Plot) (specifically, index 10).

---

## Annotation Process

The text underwent **multi-round annotation** for part-of-speech (PoS) tagging:

1. **First Round**:
   Automated tagging using **SpaCy's Danish PoS tagger** (`model="da_core_news_sm"`).

2. **Second Round**:
   **Annotator 1** manually reviewed and corrected the tags assigned in the first round.

3. **Third Round**:
   **Annotator 2** independently added their own corrected tags in parallel.

4. **Final Round**:
   **Annotator 1** compared the corrections from both annotators and **adjudicated the final tags** to resolve a few discrepancies.

---

# Dowloading the dataset
```py
# using python
from datasets import load_dataset

ds = load_dataset("chcaa/hist-dk-pos", split="train")

# if you want it as a pandas DataFrame:
df = ds.to_pandas()
```
---

## Dataset Details

This dataset indexes annotated PoS tags for historical Danish fiction.

- **Curated by:** [GoldenMatrix](https://chc.au.dk/research/golden-matrix) at Center for Humanities Computing (CHC), Aarhus University
- **Processed by:** [ENO](https://hislab.quarto.pub), Aalborg University
- **Uploaded by:** [Pascale Feldkamp](https://huggingface.co/PascaleF)
- **Language(s):** Danish (dan), from 1824
- **License:** Danish Newspapers fall under Public Domain (CC0)
  
---