Emotion Detecting Model is a machine learning project designed to detect and predict emotional turmoil in individuals. Emotional turmoil is a complex, multifaceted state, often reflecting deep psychological stress, anxiety, and fluctuating emotions. By analyzing physiological, behavioral, and contextual features, this model aims to identify and quantify these emotional states with high accuracy.
To build a robust predictive model that can understand and detect emotional turmoil using a rich dataset of physiological, behavioral, and contextual variables. This model can be used in applications such as mental health monitoring, mood-aware systems, and intelligent therapy support.
The project implements multiple supervised learning techniques to classify emotional turmoil:
- ✅ Decision Tree Classifier
- ✅ Random Forest Classifier
- ✅ XGBoost Classifier
Each model was trained, validated, and evaluated using performance metrics to select the most suitable one for final deployment.
| Technology | Description |
|---|---|
 |
Core programming language |
 |
Data manipulation |
 |
Numerical operations |
 |
Machine learning models |
 |
Visualization (optional) |
 |
Gradient boosting |
- Physiological: Heart rate, sleep patterns, hormone levels, etc.
- Behavioral: Social activity, digital interaction patterns, speech/text sentiment.
- Contextual: Environmental factors, location data, work/school schedule, etc.
These features were carefully preprocessed, normalized, and used to train the machine learning algorithms.
- The final model predictions are saved in:
📄submission.csv— Each row corresponds to an individual's emotional turmoil status.
- Feature engineering based on domain knowledge in psychology and behavioral science
- Model comparison using accuracy, precision, recall, and F1-score
- Use of ensemble learning (Random Forest, XGBoost) for improved generalization
- Interpretability through Decision Trees and feature importance analysis