amitca71/marabert2-levantine-hate-model

Levantine Hate Speech Classifier

This model identifies Hate Speech and Abusive Language in Levantine Arabic tweets. It was fine-tuned on the dataset: amitca71/marabert2-levantine-hate-model-dataset.

📊 Evaluation

Metric	Score
Accuracy	0.8403648802736602
F1 Score	0.7648624556266904

📖 Labels

ID	Label	Definition
0	Abusive	Insults/Vulgarity not directed at identity.
1	Normal	Safe text.
2	Hate	Attacks based on religion, sect, gender, or nationality.

Usage

from transformers import pipeline
classifier = pipeline("text-classification", model="amitca71/marabert2-levantine-hate-model")
classifier("يا زلمة حل عني")