Standing up for or against: A text-mining study on the recommendation of mobile payment apps

Abstract

Mobile payment systems offer enormous potential as alternative payment solutions. However, the diffusion of mobile payments over the years has been less than optimal despite the numerous studies that have explored the reasons for its adoption. Consequently, there is an increased interest in exploring alternative actions for promoting its diffusion, especially user recommendation of the technology. This is because positive recommendations can enormously influence the decisions of potential consumers to use the technology while negative recommendations can increase resistance to it. The few extant studies in this domain have followed the traditional survey approach with hypothetic-deductive reasoning, thus limiting an understanding of factors outside their conceptual models that could influence recommendations. To address this shortcoming, this study uses a qualitative text-mining approach that explores themes from user reviews of mobile payment applications (apps). Using 5955 reviews from 16 mobile payment apps hosted on the Google Play store, this study applied the latent Dirichlet allocation (LDA) text-mining method to extract themes from the reviews that help to explain why users provide positive or negative recommendations about mobile payment systems. A total of 13 themes (i.e. ease of use, usefulness, convenience, security, reliability, satisfaction, transaction speed, time-saving, customer support, output quality, perceived cost, usability and trust) were generated from the LDA model which provides both theoretical and practical insights for advancing mobile payments diffusion and research.