FastAPI-powered survey application with advanced statistical analysis.

• Survey: https://glowstone.red/SurveyAI-UIUC

• Interactive Survey: 6 AI-related confidence questions (-1 to +1 scale)
• Real-time Visualizations: Histograms with mean indicators and outlier detection
• Multivariate Regression: OLS/Ridge regression with statsmodels
• Advanced Analytics: Correlation matrices, covariance, PCA, Cronbach's Alpha, KMO
• Admin Dashboard: Data management, export, deletion, synthetic data generation

Backend: FastAPI, Statsmodels, Scikit-learn, Pandas, NumPy, Uvicorn Frontend: React 18, Chart.js, Tailwind CSS Deployment: cPanel/Passenger, Systemd, JSON storage

SurveyAI/
├── main.py                 # FastAPI backend
├── v0_2_analytics.py       # Advanced analytics
├── index.html             # Survey interface
├── requirements.txt        # Dependencies
└── docs/                  # Deployment guides

pip install -r requirements.txt
python main.py
# Access at http://localhost:8000

Deployment: See docs/QUICK_START_CPANEL.md or docs/DEPLOYMENT.md

• GET / - Survey page
• GET /api/responses - Get all responses
• POST /api/submit - Submit response
• GET /api/regression - Regression analysis
• GET /api/regression-age - Age-based models
• GET /api/analytics - Advanced analytics
• POST /api/admin/export - Export data (admin)
• POST /api/admin/clear - Clear data (admin)
• POST /api/admin/generate - Generate samples (admin)
• POST /api/admin/delete - Delete response (admin)

API docs: /docs

• SHA-256 password authentication. Default: PA$$
• Update in main.py line 37: ADMIN_PASSWORD_HASH = hashlib.sha256(b"your_password").hexdigest()

1. Humans have access to sentient AI today
2. AI outperforms humans on EQ exam by 2030
3. I can compete in job market without AI
4. Child born today faces harder path to higher education (HS 2044)
5. I understand how Multi-Modal ML/AI models decide
6. AI positively impacts my social circle by 2035

Distributions: 9 equal-width histograms, mean line, outlier highlighting (Pink: |Z| >= 2.24, Red: |Z| >= 2.89)

Regression: 6 models (question vs others + age), Beta/SE/t-stats/P-values, R²/Adj R²/RMSE, age prediction

Advanced: Correlation/Covariance matrices, PCA, Cronbach's Alpha, KMO, rolling trends, scatter plots

{
  "responses": [{
    "timestamp": "2025-10-22T12:00:00Z",
    "age_group": "23-26",
    "q1": 0.5, "q2": -0.3, "q3": 0.1,
    "q4": 0.2, "q5": -0.4, "q6": 0.6
  }]
}

~Mason 'A' "ShlinkLFO" Beydoun GLOWSTONE LLC [email protected] MIT License - UIUC MS Business Analytics Program