Deep Adaptive Feature Fusion for OriginDestination Passenger Flow Forecasting in Mass Events

This project aims to forecast passenger flow for mass events using a Deep Adaptive Feature Fusion approach. By leveraging multiple machine learning algorithms, including Random Forest, Linear Regression, Stacking Regressor, and XGBoost Regression, the system is trained to predict both inflow and outflow of passengers. The model will be trained using historical event data, considering various influencing factors such as time, location, and event type. Based on this input data, the system will predict the number of people likely to attend the event and the estimated outflow, assisting in better crowd management and planning for large-scale events.

Keywords

Mass Events, Passenger Flow, Inflow Prediction, Outflow Prediction, Random Forest, Linear Regression, Stacking Regressor, XGBoost Regression, Event Management, Predictive Modeling.