Urban Traffic Congestion Analysis Using a Hybrid Machine Learning Model: A Case Study of Nasiriyah

In many Iraqi cities, urban traffic congestion is still a problem, and the use of sophisticated analytical methods is constrained by the lack of trustworthy data.The present research combines both supervised learning and clustering algorithms to create a data-driven model to classify traffic levels in Nasiriyah.An example of K-means clustering was used to derive a categorical congestion-level variable by using field data that were collected on sixteen different districts to explain the underlying traffic patterns.The ability of three classification algorithms-J48 decision tree, Naive Bayes, and random forest (RF) to differentiate between low, medium, and high congestion circumstances was then assessed.With an accuracy of 81.25% and a kappa value of 0.70, the J48 model outperformed the other classifiers on the short dataset and had the most consistent performance.The findings also suggest that the lightweight hybrid strategy can provide authoritative congestion information in data-limited settings and, therefore, present a useful tool to support planning and traffic management decisions in fast-growing cities.