An XGBoost-based method for predicting urban transport modes in the Rabat area

A key factor driving challenges in urban development is the continuous rise in population density and the increasing average distance traveled by individuals using both public and private transportation. Private vehicles, in particular, contribute significantly to issues such as congestion, air pollution, noise, and other related urban problems. In this context, machine learning offers valuable tools for developing predictive models. To address these challenges, we proposed an XGBoost algorithm to predict transportation mode choices in urban areas. The model demonstrated strong performance, achieving 96.86% accuracy in our experiments, underscoring its potential for optimizing urban mobility planning and mitigating transportation-related issues.