mohanrj/mybully-emobert-manual

MYBully-EmoBERT (Manual)

Model Overview

This is MYBully-EmoBERT (Manual), trained solely on manual annotations of the MYBully dataset for Emotion Detection.
It serves as a baseline model.

Intended Use

• Multi-class emotion classification of tweets.

Training Data

• Dataset: MYBully (Bahasa Malaysia tweets).
• Annotation: Manual
• {'Anger': 0, 'Disgust': 1, 'Fear': 2, 'Happiness': 3, 'Neutral': 4, 'Sadness': 5, 'Surprise': 6}

Model Details

• Base model: roberta-base-bahasa-cased
• Fine-tuning: Multi-class classification head
• Labels: Multiple emotions (e.g., Anger, Joy, Sadness, Fear, Neutral)

Performance

Metric	Value
Accuracy	0.65
Precision	0.66
Recall	0.65
F1	0.65