Datasets:
annotations_creators:
- expert-generated
language_creators:
- found
languages:
- catalan
licenses:
- cc-by-sa-4.0
multilinguality:
- monolingual
pretty_name: VilaQuAD
size_categories:
- 1K<n<10K
source_datasets:
- original
task_categories:
- question-answering
task_ids:
- extractive-qa
Dataset Card for VilaQuAD
Table of Contents
- Table of Contents
- Dataset Description
- Dataset Structure
- Dataset Creation
- Considerations for Using the Data
- Additional Information
Dataset Description
- Homepage: https://doi.org/10.5281/zenodo.4562337
- Paper: Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? A Comprehensive Assessment for Catalan
- Point of Contact: Carlos Rodríguez-Penagos and Carme Armentano-Oller
Dataset Summary
VilaQuAD, An extractive QA dataset for catalan, from Vilaweb newswire text.
This dataset contains 2095 of Catalan language news articles along with 1 to 5 questions referring to each fragment (or context).
VilaQuad articles are extracted from the daily Vilaweb (www.vilaweb.cat) and used under CC-by-nc-sa-nd licence.
This dataset can be used to build extractive-QA and Language Models.
Supported Tasks and Leaderboards
Extractive-QA, Language Model.
Languages
Catalan (ca).
Dataset Structure
Data Instances
{
'id': 'P_556_C_556_Q1',
'title': "El Macba posa en qüestió l'eufòria amnèsica dels anys vuitanta a l'estat espanyol",
'context': "El Macba ha obert una nova exposició, 'Gelatina dura. Històries escamotejades dels 80', dedicada a revisar el discurs hegemònic que es va instaurar en aquella dècada a l'estat espanyol, concretament des del començament de la transició, el 1977, fins a la fita de Barcelona 92. És una mirada en clau espanyola, però també centralista, perquè més enllà dels esdeveniments ocorreguts a Catalunya i els artistes que els van combatre, pràcticament només s'hi mostren fets polítics i culturals generats des de Madrid. No es parla del País Basc, per exemple. Però, dit això, l'exposició revisa aquesta dècada de la història recent tot qüestionant un triomfalisme homogeneïtzador, que ja se sap que va arrasar una gran quantitat de sectors crítics i radicals de l'àmbit social, polític i cultural. Com diu la comissària, Teresa Grandas, de l'equip del Macba: 'El relat oficial dels anys vuitanta a l'estat espanyol va prioritzar la necessitat per damunt de la raó i va consolidar una mirada que privilegiava el futur abans que l'anàlisi del passat recent, obviant qualsevol consideració crítica respecte de la filiació amb el poder franquista.",
'question': 'Com es diu la nova exposició que ha obert el Macba?',
'answers': [
{
'text': "'Gelatina dura. Històries escamotejades dels 80'",
'answer_start': 38
}
]
}
Data Fields
Follows Rajpurkar, Pranav et al., 2016 for squad v1 datasets.
id(str): Unique ID assigned to the question.title(str): Title of the Wikipedia article.context(str): Wikipedia section text.question(str): Question.answers(list): List of answers to the question, each containing:text(str): Span text answering to the question.answer_startStarting offset of the span text answering to the question.
Data Splits
- train.json: 1295 contexts, 3882 questions
- dev.json: 400 contexts, 1200 questions
- test.json: 400 contexts, 1200 questions
Dataset Creation
Methodology
From a the online edition of the catalan newspaper Vilaweb (https://www.vilaweb.cat), 2095 articles were randomnly selected. These headlines were also used to create a Textual Entailment dataset. For the extractive QA dataset, creation of between 1 and 5 questions for each news context was commissioned, following an adaptation of the guidelines from SQUAD 1.0 (Rajpurkar, Pranav et al. “SQuAD: 100, 000+ Questions for Machine Comprehension of Text.” EMNLP (2016)), http://arxiv.org/abs/1606.05250. In total, 6282 pairs of a question and an extracted fragment that contains the answer were created.
Curation Rationale
For compatibility with similar datasets in other languages, we followed as close as possible existing curation guidelines. We also created another QA dataset with wikipedia to ensure thematic and stylistic variety.
Source Data
Initial Data Collection and Normalization
The source data are scraped articles from archives of Catalan newspaper website Vilaweb.
Who are the source language producers?
[More Information Needed]
Annotations
Annotation process
We comissioned the creation of 1 to 5 questions for each context, following an adaptation of the guidelines from SQUAD 1.0 (Rajpurkar, Pranav et al. “SQuAD: 100, 000+ Questions for Machine Comprehension of Text.” EMNLP (2016)).
Who are the annotators?
Annotation was commissioned to an specialized company that hired a team of native language speakers.
Personal and Sensitive Information
No personal or sensitive information included.
Considerations for Using the Data
Social Impact of Dataset
[More Information Needed]
Discussion of Biases
[More Information Needed]
Other Known Limitations
[More Information Needed]
Additional Information
Dataset Curators
Carlos Rodríguez-Penagos (carlos.rodriguez1@bsc.es) and Carme Armentano-Oller (carme.armentano@bsc.es)
Licensing Information
This work is licensed under a Attribution-ShareAlike 4.0 International License.
Citation Information
@inproceedings{armengol-estape-etal-2021-multilingual,
title = "Are Multilingual Models the Best Choice for Moderately Under-resourced Languages? {A} Comprehensive Assessment for {C}atalan",
author = "Armengol-Estap{\'e}, Jordi and
Carrino, Casimiro Pio and
Rodriguez-Penagos, Carlos and
de Gibert Bonet, Ona and
Armentano-Oller, Carme and
Gonzalez-Agirre, Aitor and
Melero, Maite and
Villegas, Marta",
booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-acl.437",
doi = "10.18653/v1/2021.findings-acl.437",
pages = "4933--4946",
}
Funding
This work was funded by the Catalan Ministry of the Vice-presidency, Digital Policies and Territory within the framework of the Aina project.