Browse Datasets

AfriHate

Hate Speech Detection for 15 African Languages

hate-speech-detection text-classification safety ~45,000 examples Apache 2.0
Access Dataset

Description

Multilingual dataset for hate speech and abusive language detection. Tweets annotated by native speakers with socio-cultural context.

Quick Start

Python 
from datasets import load_dataset

dataset = load_dataset("afrihate/afrihate", "hau", split="train")

Project Ideas

Hate Speech Classifier

Train BERT classifier for hate speech in Kinyarwanda

Intermediate

Feature Analysis

Investigate linguistic features correlating with abusive labels

Advanced

Ethical Considerations

  • Contains offensive content
  • Use strictly for safety research