Modeling Intention and Frequency of Cashless Behavior: Integrating Big Data and Cluster-Structural Equations Modeling on Bank Customers

This study aims to determine and model the determinant factors that increase the intensity and frequency of use of cashless payments in the community among Bank customers in Indonesia. The novelty in this research is to combine big data methods (web scraping & LDA) and statistical modeling (cluster and SEM) in modeling the intensity and frequency of use of cashless society. This research uses research data obtained from questionnaires distributed using indicators obtained using big data methods and modeled using cluster-SEM. The results show that there are three groups of cashless payment usage levels, the three groups produce three SEM models where each group has different determinant factors in modeling the intensity and frequency of cashless usage. For the public to increase their sensitivity to the use of a cashless society because its development is very fast. For banks to maximize economic, payment, adoption, policy, and consumer focus variables which can increase the intensity and frequency of use of a cashless society.