Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,28 @@
|
|
| 1 |
-
|
| 2 |
-
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Kazakh Swear Words Dataset 🇰🇿
|
| 2 |
+
|
| 3 |
+
Dataset of Kazakh obscene and profane expressions for NLP tasks:
|
| 4 |
+
- text classification
|
| 5 |
+
- moderation
|
| 6 |
+
- toxicity detection
|
| 7 |
+
- LLM fine-tuning
|
| 8 |
+
|
| 9 |
+
## Languages
|
| 10 |
+
- Kazakh (`kk`)
|
| 11 |
+
|
| 12 |
+
## Labels
|
| 13 |
+
- `1` — obscene / profanity
|
| 14 |
+
- `0` — neutral
|
| 15 |
+
|
| 16 |
+
## Fields
|
| 17 |
+
- `text` — raw text
|
| 18 |
+
- `label` — class
|
| 19 |
+
- `category` — type of profanity
|
| 20 |
+
- `language` — ISO 639-1 code
|
| 21 |
+
|
| 22 |
+
## Use cases
|
| 23 |
+
- content moderation
|
| 24 |
+
- chatbot filtering
|
| 25 |
+
- dataset for low-resource languages
|
| 26 |
+
|
| 27 |
+
## Disclaimer
|
| 28 |
+
This dataset contains offensive language and is intended for research purposes only.
|