Dubai Market Demand Intelligence System

A data-driven system that forecasts sector-specific economic activity in Dubai using synthetic data (mimicking real-world signals) and advanced time-series forecasting.

Features

• Multi-Sector Analysis: Real Estate, Tourism, Logistics, Retail, and Technology.
• Composite Demand Index: 0-100 scale indicating the "heat" of each sector.
• Forecasting: 90-day projections using Exponential Smoothing (Holt-Winters).
• Interactive Dashboard: Built with Streamlit and Plotly for real-time exploration.

Project Structure

.
├── app/                    # Streamlit Application
│   ├── main.py             # Entry point
│   └── pages/              # Deep dive pages
├── data/                   # Data storage
├── src/                    # Core logic
│   ├── ingestion/          # Data loaders (Synthetic + Real)
│   ├── processing/         # Index calculation logic
│   └── forecasting/        # Forecasting models
└── requirements.txt        # Dependencies

Setup & Installation

1. Clone the repository:

   git clone <repo_url>
   cd dubai_market_intelligence

2. Create a virtual environment:

   python3 -m venv venv
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the Dashboard:

   export PYTHONPATH=$PYTHONPATH:.
   streamlit run app/main.py

Data Strategy

Currently, the system uses a Synthetic Data Generator (src/ingestion/loader.py) to create realistic historical data (seasonality, trends, noise) for demonstration purposes.

To use real data:

1. Place CSV files in data/raw/.
2. Implement a new loader in src/ingestion/ inheriting from DataLoader.
3. Update app/main.py to use the new loader.