Technical Project: Remote Work Burnout Analysis & Prediction

Note

This project was developed via Antigravity AI.
🎓 Read the Senior Evaluation Report (Audit) - A multi-perspective analysis (Junior vs. Senior vs. Recruiter). The analysis is based on Synthetic Data to ensure realistic behavioral modeling.

🚀 Overview

This repository contains a test and development-focused analysis and predictive modeling project focused on occupational health in remote work environments. The project is based on the Work From Home Employee Burnout Dataset from Kaggle.

📊 Key Findings

• Sleep Deprivation & Burnout: Sleep duration is the strongest inversely correlated factor with burnout risk.
• The Screen Intensity Threshold: Employees exceeding 9 hours of screen time show a disproportionate increase in burnout scores.
• Productivity Paradox: High task completion rates do not necessarily indicate well-being; they often precede a "High Risk" burnout phase when coupled with low sleep.

🛠️ Tech Stack

• Analysis: Pandas, NumPy, SciPy
• Visualization: Seaborn, Matplotlib, Plotly (Interactive EDA)
• Machine Learning: XGBoost, LightGBM, Scikit-Learn
• Explainable AI (XAI): SHAP (TreeExplainer)

🏗️ Methodological Rigor

Acknowledging the Class Imbalance (High Risk at ~5%), the following senior-level mitigations were implemented:

• Stratified Splitting: Ensuring representative samples in both training and validation sets.
• Recall (Sensitivity) Focus: Since the cost of missing a high-risk employee (False Negative) is high, the model evaluation prioritizes Recall for the 'High' class over simple accuracy.
• Advanced Ratios: Utilizing behavioral indexes like Sleep Efficiency to sharpen decision boundaries for minority classes.

Tip

AI as a Strategic Senior Partner: Beyond code generation, Antigravity AI was utilized in this project as a "Senior Evaluator." It provided multi-perspective feedback (Junior vs. Senior vs. Interviewer) and strategic advice, showcasing an unconventional use of AI as a high-level consultant for project auditing and career-oriented refinement.

💻 Installation & Usage

1. Clone the repository.
2. Install dependencies:

   pip install -r requirements.txt

3. Run the Jupyter Notebook:

   jupyter notebook Burnout_Analysis.ipynb

📝 License

MIT License. Feel free to use and contribute.