Skip to content

Dataset · uitnlp

Annotated Vietnamese comments for hate speech detection

uitnlp/vihsd

A corpus of 33 thousand Vietnamese social media comments for training content moderation.

What's inside

The dataset includes about 33 thousand Vietnamese social media text comments. Each comment is manually assigned to one of three classes: HATE, OFFENSIVE, or CLEAN. The set is compatible with the standard datasets library loader.

What you can do with it

  • Training classifiers for automated toxic content moderation in Vietnamese
  • Benchmarking multilingual models in hate speech detection tasks
  • Fine-tuning safety filters for chatbots working with Vietnamese audience
Tasks
Text classification
Languages
Vietnamese
Size
10K–100K rows
Downloads this month
338
Likes
12
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("uitnlp/vihsd")