Emotional Artificial Neural Networks and Gaussian Process-Regression-Based Hybrid Machine-Learning Model for Prediction of Security and Privacy Effects on M-Banking Attractiveness

【Author】 Cavus, Nadire; Mohammed, Yakubu Bala; Gital, Abdulsalam Ya'u; Bulama, Mohammed; Tukur, Adamu Muhammad; Mohammed, Danlami; Isah, Muhammad Lamir; Hassan, Abba

【Source】SUSTAINABILITY

【影响因子】3.889

【Abstract】With recent advances in mobile and internet technologies, the digital payment market is an increasingly integral part of people's lives, offering many useful and interesting services, e.g., m-banking and cryptocurrency. The m-banking system allows users to pay for goods, services, and earn money via cryptotrading using any device such as mobile phones from anywhere. With the recent trends in global digital markets, especially the cryptocurrency market, m-banking is projected to have a brighter future. However, information stored or conveyed via these channels is more vulnerable to different security threats. Thus, the aim of this study is to examine the influence of security and confidentiality on m-banking patronage using artificial intelligence ensemble methods (ANFIS, GPR, EANN, and BRT) for the prediction of safety and secrecy effects. AI models were trained and tested using 745 datasets obtained from the study areas. The results indicated that AI models predicted the influence of security with high precision (NSE > 0.95), with the GPR model outperformed the other models. The results indicated that security and privacy were key influential parameters of m-payment system patronage (m-banking), followed by service and interface qualities. Unlike previous m-banking studies, the study results showed ease of use and culture to have no influence on m-banking patronage. These study results would assist m-payment system stakeholders, while the approach may serve as motivation for researchers to use AI techniques. The study also provides directions for future m-banking studies.

【Keywords】m-banking; security; artificial intelligence; ensemble techniques; machine learning; privacy

【发表时间】2022 MAY

【收录时间】2022-06-10

【文献类型】实证性文章

【主题类别】

区块链治理-技术治理-