A Comprehensive Analysis of Clustering Public Utility Bus Passenger’s Behavior during the COVID-19 Pandemic: Utilization of Machine Learning with Metaheuristic Algorithm

Public utility bus (PUB) systems and passenger behaviors drastically changed during the COVID-19 pandemic. This study assessed clustered behavior of 505 PUB passengers using feature selection, K-means clustering, particle swarm optimization (PSO). The wrapper method was seen to be best among six selection techniques through recursive with a 90% training set 10% testing set. It revealed that this technique produced 26 optimal subsets. These features were then fed into clustering PSO find passengers’ clusters. algorithm tested 12 different parameter settings outcome. As result, combination 23 Utilizing Pareto analysis, only considered vital Specifically, five clusters found have comprehensive similarities in demographics responses. stakeholders could use cluster findings as benchmark improve current system.