Skip to content

README.md

Latest commit

 

History

History
293 lines (206 loc) · 8.55 KB

README.md

File metadata and controls

293 lines (206 loc) · 8.55 KB

🧠 Advanced Depression Predictor Model

Python XGBoost Streamlit License

A machine learning project for depression prediction using synthetic data and XGBoost classification

🚀 Features • 📊 Dataset • ⚙️ Installation • 🎯 Usage • 📈 Model Performance

📖 Overview

This project represents an advanced, unbiased depression prediction system built as a continuation of a previous model trained on noisy data. The current implementation leverages a synthetic, ideal, and unbiased dataset to ensure clean experimentation and demonstrate high-performance machine learning capabilities.

🎯 Objectives

  • Build a robust binary classification model for depression prediction
  • Demonstrate clean ML experimentation using synthetic, bias-free data
  • Provide an interactive interface for real-time mental health predictions
  • Showcase feature importance and model interpretability

⚠️ Disclaimer: This project is for experimental and educational purposes only. It does not replace professional medical diagnosis or clinical assessment.

✨ Features

🔬 Machine Learning

  • XGBoost Classifier for high-performance binary classification
  • TF-IDF Vectorizer for text preprocessing during training
  • 7 key psychological and behavioral features:
    • sadness_score - Intensity of sadness feelings
    • anxiety_score - Level of anxiety experienced
    • fatigue_score - Physical and mental exhaustion
    • sleep_issues - Sleep quality and disturbances
    • social_withdrawal - Social isolation tendencies
    • concentration_issues - Difficulty focusing
    • mood_swings - Emotional volatility

📊 Visualization & Analysis

  • Correlation heatmaps for feature relationships
  • Feature importance graphs
  • Distribution plots for numeric features
  • Prediction confidence metrics
  • Interactive data exploration

💻 Interactive Application

  • Streamlit-powered web interface
  • Real-time predictions with slider inputs
  • Visual feedback and charts
  • User-friendly design with immediate results

📊 Dataset

Characteristics

  • Size: 10,000 samples
  • Features: 7 numeric features (continuous scores and boolean indicators)
  • Target: Binary classification (0 = Not Depressed, 1 = Depressed)
  • Quality: Synthetic, ideal, and unbiased for controlled testing
  • Format: CSV file (depression_dataset_10000.csv)

Feature Distribution

All features are carefully balanced to ensure unbiased learning and testing of prediction logic in a controlled environment.

🛠️ Tech Stack

Category Technologies
Core Language Python 3.8+
ML Framework XGBoost, Scikit-learn
Data Processing Pandas, NumPy
Visualization Matplotlib, Seaborn
Web Interface Streamlit
Model Persistence Joblib

⚙️ Installation

Prerequisites

  • Python 3.8 or higher
  • pip package manager
  • Virtual environment (recommended)

Setup Instructions

  1. Clone the repository

    git clone https://github.com/willow788/Advanced-depression-predictor-model.git
cd Advanced-depression-predictor-model

  2. Create a virtual environment (recommended)

    # Windows
python -m venv venv
venv\Scripts\activate

# macOS/Linux
python3 -m venv venv
source venv/bin/activate

  3. Install dependencies

    pip install -r requirements.txt

    If requirements.txt doesn't exist, install manually:

    pip install xgboost pandas numpy matplotlib seaborn scikit-learn streamlit joblib jupyter

🎯 Usage

🔧 Model Training

To retrain the model or explore the training process:

jupyter notebook prediction.ipynb

The notebook includes:

  • Data loading and exploration
  • Feature engineering
  • Model training with XGBoost
  • Performance evaluation
  • Visualization of results

🚀 Running the Streamlit App

Launch the interactive web application:

streamlit run app.py

The app will open in your default browser at http://localhost:8501

🎮 Using the Application

  1. Input Parameters: Use sliders and checkboxes to set mental health indicators
  2. Generate Prediction: Click to analyze depression likelihood
  3. View Results: See prediction, confidence score, and visualizations
  4. Explore Charts: Analyze feature importance and distributions

📈 Model Performance

Accuracy Metrics

  • Training Accuracy: 100%
  • Test Accuracy: 100%

📌 Note: Perfect accuracy is achieved due to the synthetic, idealized nature of the dataset. This represents a controlled testing environment, not real-world deployment readiness.

Model Artifacts

File Description Size
depression_xgb_model.pkl Trained XGBoost classifier ~218 KB
tfidf_vectorizer.pkl TF-IDF vectorizer for text features ~3.6 KB

Both models are serialized using joblib and loaded during inference.

📁 Project Structure

Advanced-depression-predictor-model/
│
├── app.py                          # Streamlit application
├── prediction. ipynb                # Training notebook
├── depression_dataset_10000.csv    # Synthetic dataset
├── depression_xgb_model.pkl        # Trained model
├── tfidf_vectorizer.pkl            # TF-IDF vectorizer
├── Screenshot 2025-10-15 022315.png # App interface demo
└── README.md                       # Documentation

🖼️ Screenshots

Application Interface

App Interface

The Enhanced Depression Prediction App allows users to input psychological parameters through an intuitive interface and receive instant predictions with visual analytics.

🔬 Methodology

Training Pipeline

  1. Data Generation: Synthetic dataset creation with controlled distributions
  2. Feature Engineering: Numerical and text feature processing
  3. Model Selection: XGBoost chosen for gradient boosting capabilities
  4. Training: Supervised learning with binary classification
  5. Evaluation: Performance metrics and visualization
  6. Deployment: Model serialization and Streamlit integration

Key Design Decisions

  • XGBoost: Selected for handling non-linear relationships and feature importance
  • Synthetic Data: Ensures bias-free testing environment
  • Interactive UI: Streamlit provides rapid prototyping and user engagement

🚧 Future Enhancements

  • Add SHAP values for model explainability
  • Implement ensemble methods (Random Forest, Neural Networks)
  • Create API endpoint for programmatic access
  • Add unit tests and CI/CD pipeline
  • Deploy to cloud platform (Streamlit Cloud, Heroku)
  • Include multi-class severity classification
  • Add data validation and input sanitization

📝 Important Notes

⚠️ Limitations

  • Synthetic Data: Results do not represent real-world clinical accuracy
  • Educational Purpose: Not validated for medical or diagnostic use
  • Bias Considerations: Real-world depression involves complex factors not captured here
  • No Clinical Validation: Not approved by medical authorities

🔒 Ethical Considerations

  • This tool should never replace professional mental health assessment
  • Predictions are based on simplified synthetic patterns
  • Always consult qualified healthcare professionals for mental health concerns

🤝 Contributing

Contributions are welcome! Feel free to:

  • Report bugs or issues
  • Suggest new features
  • Submit pull requests
  • Improve documentation

📄 License

This project is available for educational and research purposes. Please use responsibly and ethically.

👤 Author

willow788

🙏 Acknowledgments

  • XGBoost community for the powerful ML framework
  • Streamlit team for the intuitive app framework
  • Open-source contributors in the ML/mental health space

📞 Contact & Support

If you have questions or feedback:

  • Open an issue on GitHub
  • Star ⭐ the repository if you found it helpful!

Built with ❤️ for machine learning education