Datasets:
Tasks:
Text Classification
Formats:
csv
Languages:
Portuguese
Size:
10K - 100K
Tags:
hate-speech-detection
License:
annotations_creators: | |
- crowdsourced | |
language_creators: | |
- crowdsourced | |
language: | |
- pt | |
license: | |
- cc-by-sa-4.0 | |
multilinguality: | |
- monolingual | |
size_categories: | |
- 1K<n<10K | |
source_datasets: | |
- original | |
task_categories: | |
- text-classification | |
task_ids: [] | |
pretty_name: TuPy | |
language_bcp47: | |
- pt-BR | |
tags: | |
- hate-speech-detection | |
configs: | |
- config_name: multi-label | |
data_files: | |
- split: full | |
path: tupy_dummy_vote.csv | |
# Portuguese Hate Speech Dataset (TuPy) | |
The Portuguese hate speech dataset (TuPy) is an annotated corpus designed to facilitate the development of advanced hate speech detection models using machine learning (ML) and natural language processing (NLP) techniques. TuPy is formed by 10000 thousand unpublished annotated tweets collected in 2023. | |
This repository is organized as follows: | |
```sh | |
root. | |
├── annotations : classification given by annotators | |
├── raw corpus : dataset before being split between annotators | |
├── tupy datasets : combined result of annotations | |
└── README.md | |
``` | |
## Voting process | |
To generate the binary matrices, we employed a straightforward voting process. Three distinct evaluations were assigned to each document. In cases where a document received two or more identical classifications, the adopted value is set to 1; otherwise, it is marked as 0. | |
## Languages | |
The language used in the dataset is Brazilian Portuguese. The associated BCP-47 code is pt-BR. | |
# Dataset Structure | |
## Data Instances | |
A data point comprises the tweet text (a string) along with thirteen categories, each category is assigned a value of 0 when there is an absence of aggressive or hateful content and a value of 1 when such content is present. These values represent the consensus of annotators regarding the presence of aggressive, hate, ageism, aporophobia, body shame, capacitism, lgbtphobia, political, racism, religious intolerance, misogyny, xenophobia, and others. | |
An illustration from the multilabel ToLD-Br dataset is depicted below: | |
``` | |
{'text': 'e tem pobre de direita imbecil que ainda defendia a manutenção da política de preços atrelada ao dólar link' | |
'aggressive': 1 | |
'hate': 1 | |
'ageism': 0 | |
'aporophobia': 1 | |
'body shame': 0 | |
'capacitism': 0 | |
'lgbtphobia': 0 | |
'political': 1 | |
'racism' : 0 | |
'religious intolerance' : 0 | |
'misogyny' : 0 | |
'xenophobia' : 0 | |
'other' : 0 | |
} | |
``` | |
## Data Fields | |
- Text: A string representing a anonymized tweet posted by a user. . | |
- aggressive: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits aggressive language. | |
- hate: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet is hateful. | |
- ageism: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits ageism. | |
- aporophobia: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits aporophobia. | |
- body shame: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits body shame. | |
- capacitism: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits capacitism. | |
- lgbtphobia: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits lgbtphobia. | |
- political: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits political. | |
- racism: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits racism. | |
- religious intolerance: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits religious intolerance. | |
- misogyny: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits misogyny. | |
- xenophobia: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits xenophobia. | |
- other: Binary values (0 or 1) representing the consensus among annotators on whether the respective tweet exhibits other. | |
## Acknowledge | |
The TuPy project is the result of the development of Felipe Oliveira's thesis and the work of several collaborators. This project is financed by the Federal University of Rio de Janeiro ([UFRJ](https://ufrj.br/)) and the Alberto Luiz Coimbra Institute for Postgraduate Studies and Research in Engineering ([COPPE](https://coppe.ufrj.br/)). |