Emotion Shift Detector 🧠📈
EmotionShiftDetector is an AI system that detects subtle emotional transitions across conversations, not just isolated sentiment in individual messages.
Instead of answering “What emotion is this message?”, it answers:
“How is the emotion changing over time?”
The model is designed for conversation analysis, mental-health tools, customer support analytics, and emotion-aware AI agents.
🔍 What Problem Does It Solve?
Traditional sentiment analysis treats messages independently.
EmotionShiftDetector focuses on emotional dynamics by identifying:
- Emotional escalation or de-escalation
- Mood shifts across dialogue turns
- Sudden emotional changes that indicate risk, frustration, or recovery
✨ Key Features
- 🧠 Emotion classification per conversation turn
- 🔄 Detection of emotional shifts over time
- 🗣️ Human-readable explanations for each shift
- 🧩 Modular and explainable design
- 🤗 Hugging Face–ready pipeline
- 🎛️ Gradio demo included
- 🧪 Test-covered core components
📂 Project Structure
emotion-shift-detector/
├── config/
├── data/
├── src/
├── training/
├── pipelines/
├── scripts/
├── tests/
├── notebooks/
├── app.py
├── README.md
├── model_card.md
├── requirements.txt
└── LICENSE
⚙️ Installation
pip install -r requirements.txt
🚀 Quick Usage
from src.inference import EmotionShiftPipeline
pipeline = EmotionShiftPipeline()
conversation = [
"I am feeling okay today",
"This is getting frustrating",
"Now I am really angry"
]
result = pipeline(conversation)
print(result)
🎛️ Gradio Demo
python app.py
🧠 How It Works
- Text Preprocessing
- Emotion Classification
- Shift Detection
- Explanation Engine