-
Project OverviewThis data science project developed a bike-sharing demand prediction system that combines weather forecasting with machine learning to estimate bicycle rental demand across major global cities. The system helps optimize bike-sharing operations by predicting demand based on weather conditions.
-
Key Components
- Weather data integration using OpenWeather API for 5-day forecasts
- Machine learning model trained on Seoul bike-sharing historical data
- Interactive R Shiny dashboard with real-time predictions for 5 major cities
- Visualization of weather impacts on predicted demand
An R Shiny-based dashboard is designed to display the Temperature Forecast, Predicted Bike Demand Forecast for the next 5 days, and a correlation plot between Humidity and Bike Demand.
You can access the Shiny app for the final project here.