Customer Feedback Emotion & Sentiment Classifier
This repository hosts a high-performance TF-IDF + Logistic Regression text classification model trained on the dair-ai/emotion dataset.
Performance & Metrics
- Accuracy: ~86.25%
- Emotions: Joy, Sadness, Love, Anger, Fear, Surprise
Quick Python Inference
import joblib
# Download and load model directly
pipeline = joblib.load("model_pipeline.joblib")
labels = joblib.load("labels.joblib")
text = "I am so happy with this service!"
pred_idx = pipeline.predict([text])[0]
print("Predicted Emotion:", labels[pred_idx])