Enhancing public transportation through user feedback and Support Vector Machine (SVM)

Retsepile Kalaoane^*Trynos Gumbo

• University of Johannesburg, Johannesburg, South Africa

Globally, transportation systems are under pressure to adopt and ensure more sustainable, inclusive, and efficient systems as cities become more complex and connected. Consequently, in the quest to develop a functional, sustainable, and inclusive transportation system, it is mandatory to first determine the key characteristics that set apart various public transportation options. Distinguishing between modes of transportation enables a detailed examination beyond superficial characteristics, resulting in better system optimization, evidence-based and better-informed decision-making.Moreover, accurate classification of public transportation data is essential for comprehending mobility trends, achieving more guided urban planning, and assisting in the creation of intelligent transportation systems (ITS). This paper presents the application of Support Vector Machines (SVM) as a robust and effective machine-learning approach for classifying different modes of public transport. Using data collected from surveys, results show that SVM performs better than conventional classifiers like decision trees and k-nearest neighbours in terms of precision, recall, and overall accuracy. SVM models were trained and evaluated for the classification of minibus taxis, buses, and ridehailing to find the main attributes that distinguish two modes of transport from each other based on users' evaluation of their performance. The SVM model achieved a high accuracy of 90% in validation, demonstrating the effectiveness of the adopted approach.The findings reveal underlying factors that influence mobility trends, patterns, and behaviours. Ultimately, the insights derived from the SVM model provide evidence-based recommendations for the development of future sustainable public transport in African cities. They also inform better practices that empower planners to design a customer-centric transportation system, that are better suited to fulfil the desires and requirements of the community.