Datasets:
Tasks:
Text Classification
Sub-tasks:
hate-speech-detection
Languages:
Portuguese
Size:
1K<n<10K
Tags:
instagram
DOI:
Merge branch 'main' of https://huggingface.co/datasets/ruanchaves/hatebr into main
Browse files
README.md
CHANGED
|
@@ -1,11 +1,24 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
-
-
|
| 4 |
language:
|
| 5 |
- pt
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
pretty_name: HateBR - Offensive Language and Hate Speech Dataset in Brazilian Portuguese
|
| 7 |
size_categories:
|
| 8 |
- 1K<n<10K
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
---
|
| 10 |
# Dataset Card for HateBR - Offensive Language and Hate Speech Dataset in Brazilian Portuguese
|
| 11 |
|
|
@@ -23,7 +36,7 @@ HateBR is the first large-scale expert annotated corpus of Brazilian Instagram c
|
|
| 23 |
|
| 24 |
### Supported Tasks and Leaderboards
|
| 25 |
|
| 26 |
-
Hate Speech Detection
|
| 27 |
|
| 28 |
### Languages
|
| 29 |
|
|
@@ -35,7 +48,7 @@ Portuguese
|
|
| 35 |
|
| 36 |
```
|
| 37 |
{'instagram_comments': 'Plano da esquerda foi é e será sempre críticar não nunca teve um plano de governo único plano era como roubar .',
|
| 38 |
-
'offensive_language':
|
| 39 |
'offensiveness_levels': 2,
|
| 40 |
'antisemitism': False,
|
| 41 |
'apology_for_the_dictatorship': False,
|
|
@@ -53,19 +66,19 @@ Portuguese
|
|
| 53 |
|
| 54 |
### Data Fields
|
| 55 |
|
| 56 |
-
* instagram_comments
|
| 57 |
-
* offensive_language
|
| 58 |
-
* offensiveness_levels
|
| 59 |
-
* antisemitism
|
| 60 |
-
* apology_for_the_dictatorship
|
| 61 |
-
* fatphobia
|
| 62 |
-
* homophobia
|
| 63 |
-
* partyism
|
| 64 |
-
* racism
|
| 65 |
-
* religious_intolerance
|
| 66 |
-
* sexism
|
| 67 |
-
* xenophobia
|
| 68 |
-
* offensive_&_no-hate_speech
|
| 69 |
|
| 70 |
### Data Splits
|
| 71 |
|
|
@@ -81,4 +94,8 @@ No standard data splits have been provided by the authors.
|
|
| 81 |
pages={7174--7183},
|
| 82 |
year={2022}
|
| 83 |
}
|
| 84 |
-
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
annotations_creators:
|
| 3 |
+
- expert-generated
|
| 4 |
language:
|
| 5 |
- pt
|
| 6 |
+
language_creators:
|
| 7 |
+
- found
|
| 8 |
+
license: []
|
| 9 |
+
multilinguality:
|
| 10 |
+
- monolingual
|
| 11 |
pretty_name: HateBR - Offensive Language and Hate Speech Dataset in Brazilian Portuguese
|
| 12 |
size_categories:
|
| 13 |
- 1K<n<10K
|
| 14 |
+
source_datasets:
|
| 15 |
+
- original
|
| 16 |
+
tags:
|
| 17 |
+
- instagram
|
| 18 |
+
task_categories:
|
| 19 |
+
- text-classification
|
| 20 |
+
task_ids:
|
| 21 |
+
- hate-speech-detection
|
| 22 |
---
|
| 23 |
# Dataset Card for HateBR - Offensive Language and Hate Speech Dataset in Brazilian Portuguese
|
| 24 |
|
|
|
|
| 36 |
|
| 37 |
### Supported Tasks and Leaderboards
|
| 38 |
|
| 39 |
+
Hate Speech Detection
|
| 40 |
|
| 41 |
### Languages
|
| 42 |
|
|
|
|
| 48 |
|
| 49 |
```
|
| 50 |
{'instagram_comments': 'Plano da esquerda foi é e será sempre críticar não nunca teve um plano de governo único plano era como roubar .',
|
| 51 |
+
'offensive_language': True,
|
| 52 |
'offensiveness_levels': 2,
|
| 53 |
'antisemitism': False,
|
| 54 |
'apology_for_the_dictatorship': False,
|
|
|
|
| 66 |
|
| 67 |
### Data Fields
|
| 68 |
|
| 69 |
+
* **instagram_comments**: Instagram comments.
|
| 70 |
+
* **offensive_language**: A classification of comments as either offensive (True) or non-offensive (False).
|
| 71 |
+
* **offensiveness_levels**: A classification of comments based on their level of offensiveness, including highly offensive (3), moderately offensive (2), slightly offensive (1) and non-offensive (0).
|
| 72 |
+
* **antisemitism**: A classification of whether or not the comment contains antisemitic language.
|
| 73 |
+
* **apology_for_the_dictatorship**: A classification of whether or not the comment praises the military dictatorship period in Brazil.
|
| 74 |
+
* **fatphobia**: A classification of whether or not the comment contains language that promotes fatphobia.
|
| 75 |
+
* **homophobia**: A classification of whether or not the comment contains language that promotes homophobia.
|
| 76 |
+
* **partyism**: A classification of whether or not the comment contains language that promotes partyism.
|
| 77 |
+
* **racism**: A classification of whether or not the comment contains racist language.
|
| 78 |
+
* **religious_intolerance**: A classification of whether or not the comment contains language that promotes religious intolerance.
|
| 79 |
+
* **sexism**: A classification of whether or not the comment contains sexist language.
|
| 80 |
+
* **xenophobia**: A classification of whether or not the comment contains language that promotes xenophobia.
|
| 81 |
+
* **offensive_&_no-hate_speech**: A classification of whether or not the comment is offensive but does not contain hate speech.
|
| 82 |
|
| 83 |
### Data Splits
|
| 84 |
|
|
|
|
| 94 |
pages={7174--7183},
|
| 95 |
year={2022}
|
| 96 |
}
|
| 97 |
+
```
|
| 98 |
+
|
| 99 |
+
### Contributions
|
| 100 |
+
|
| 101 |
+
Thanks to [@ruanchaves](https://github.com/ruanchaves) for adding this dataset.
|