Unveiling Mobile Wallet Adoption Patterns AmongKelantanese B40: An Approach of Machine Learning

Abstract

Mobile wallet adoption among the B40 population in Kelantan, Malaysia, presents a unique challenge despite the significant attention mobile wallets have gained as a convenient and secure payment method, especially during the COVID-19 pandemic. This paper explores the application of unsupervised machine learning models to understand and address the low adoption rates of mobile wallets among this population segment. By leveraging unsupervised machine learning techniques, researchers aim to uncover patterns and relationships within the dataset related to factors influencing mobile wallet adoption. The study utilizes data collected on income level, education, employment status, geographical location, perceived usefulness, subjective norms, financial costs, data privacy, security, trust, awareness, risk, technology skills, complexity, relative convenience, and relative advantage among the Kelantanese B40 population. Through clustering algorithms, such as the self-organizing map algorithm, insights into mobile wallet adoption behavior and preferences are gained, enabling the development of targeted strategies to promote adoption and optimize mobile wallet services for this specific demographic. The study underscores the importance of government support, financial literacy programs, improved infrastructure, and user-friendly platforms in fostering mobile wallet adoption among the Kelantanese B40 population during the COVID-19 pandemic. This research contributes to the broader goal of financial inclusion and the transition towards a cashless society in Malaysia, with implications for similar economies facing similar challenges. Future research opportunities include exploring the long-term impact of mobile wallet adoption and incorporating perspectives from merchants and businesses in the adoption process.