Datasets:
task_categories:
- text-generation
language:
- eu
- es
- en
pretty_name: conan_eus
size_categories:
- 10K<n<100K
CONAN-EUS: CONAN dataset in Spanish and Basque
Dataset description
CONAN-EUS is based on the CONAN (COunter NArratives through Nichesourcing) dataset, originally provided by Chung et al., 2019. The original dataset provides Hate Speech (HS) - Counter Narrative (CN) pairs on Islamophobia in English, French and Italian.
CONAN-EUS was created by translating the English portion of the original CONAN into Basque and Spanish. The English portion in CONAN provides 6654 HS-CN pairs, which were collected through three different methods: (i) 1288 were collected through HS-CN nichesourcing sessions; (ii) 2576 were obtained in a posterior data augmentation process; (iii) 2790 were obtained through the translation of French and Italian pairs into English.
With the aim of providing data for CN generation in languages other than English, the English data in CONAN has been Machine Translated (MT) to Spanish and Basque, as well as manually post-edited.
The following table includes a quantitative description of the MT and post-edited datasets, including the division between train and test sets.
Data source | Total HS-CN count | Train/test division |
---|---|---|
Original | 1288 | Test set |
Augmented + Translated | 5366 | Train set |
Download
These four datasets can be found in this repository:
- es_MT: Spanish MT dataset
- es_post: Spanish post-edited dataset
- eu_MT: Basque MT dataset
- eu_post: Basque post-edited dataset
Each dataset is divided into train and test set, as previously stated. They can be found in this zip file.
HS-CN example
The CONAN dataset includes HS-CN pairs, regarding the topic of Muslims and Islam. An example of a HS-CN pair is illustrated below:
HS | CN |
---|---|
Islamic are criminals: they rape, enslave and murder people. Islam is more a worship than a religion and we do not have anything to share with them. | The myth that Muslims are dangerous and violent is a product of our vilifying media. Don’t believe everything you read. |
Citation
@inproceedings{bengoetxea-et-al-2024,
title={Basque and Spanish Counter Narrative Generation: Data Creation and Evaluation},
author={Jaione Bengoetxea and Yi-Ling Chung and Marco Guerini and Rodrigo Agerri},
year={in press},
publisher = "Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING)",
url = ""
}