clarin-pl/PUGG_KBQA

---
annotations_creators:
- expert-generated
language_creators: []
language:
- pl
license:
- cc-by-sa-4.0
multilinguality:
- monolingual
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- question-answering
task_ids:
- open-domain-qa
pretty_name: 'PUGG: KBQA dataset for Polish'
tags:
- knowledge graph
- KBQA
- wikipedia
- wikidata
configs:
- config_name: all
data_files:
- split: train
path: '*/train.jsonl'
- split: test
path: '*/test.jsonl'
default: true
- config_name: natural
data_files:
- split: train
path: natural/train.jsonl
- split: test
path: natural/test.jsonl
- config_name: template-based
data_files:
- split: train
path: template-based/train.jsonl
- split: test
path: template-based/test.jsonl
---
# PUGG: KBQA, MRC, IR Dataset for Polish

## Description

This repository contains the PUGG dataset designed for three NLP tasks in the Polish language:

- KBQA (Knowledge Base Question Answering)
- MRC (Machine Reading Comprehension)
- IR (Information Retrieval)

## Paper

For more detailed information, please refer to our research paper titled:

**"Developing PUGG for Polish: A Modern Approach to KBQA, MRC, and IR Dataset Construction"**

Authored by:
* Albert Sawczyn
* Katsiaryna Viarenich
* Konrad Wojtasik
* Aleksandra Domogała
* Marcin Oleksy
* Maciej Piasecki
* Tomasz Kajdanowicz

**The paper was accepted for ACL 2024 (findings).**

## Repositories

The dataset is available in the following repositories:

* [General](https://huggingface.co/datasets/clarin-pl/PUGG) - contains all tasks (KBQA, MRC, IR*)

For more straightforward usage, the tasks are also available in separate repositories:

* [KBQA](https://huggingface.co/datasets/clarin-pl/PUGG_KBQA) **(this repository)**
* [MRC](https://huggingface.co/datasets/clarin-pl/PUGG_MRC)
* [IR](https://huggingface.co/datasets/clarin-pl/PUGG_IR)

The knowledge graph for KBQA task is available in the following repository:

* [Knowledge Graph](https://huggingface.co/datasets/clarin-pl/PUGG_KG)

Note: If you want to utilize the IR task in the BEIR format (`qrels` in `.tsv` format), please
download the [IR](https://huggingface.co/datasets/clarin-pl/PUGG_IR) repository.

## Links

* Code:
* [Github](https://github.com/CLARIN-PL/PUGG)
* Paper:
* ACL - TBA
* [Arxiv](https://arxiv.org/abs/2408.02337)

## Citation

```bibtex
@misc{sawczyn2024developingpuggpolishmodern,
title={Developing PUGG for Polish: A Modern Approach to KBQA, MRC, and IR Dataset Construction},
author={Albert Sawczyn and Katsiaryna Viarenich and Konrad Wojtasik and Aleksandra Domogała and Marcin Oleksy and Maciej Piasecki and Tomasz Kajdanowicz},
year={2024},
eprint={2408.02337},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2408.02337},
}
```

## Contact

[e-mail]

## Usage

```python
from datasets import load_dataset

# loading all

dataset = load_dataset("clarin-pl/PUGG_KBQA")
# or
dataset = load_dataset("clarin-pl/PUGG_KBQA", "all")

print(dataset)

# loading natural
dataset = load_dataset("clarin-pl/PUGG_KBQA", "natural")
print(dataset)

# loading template-based
dataset = load_dataset("clarin-pl/PUGG_KBQA", "template-based")
print(dataset)
```