A Hybrid SEM and Neural Network Approach to Understand and Predict the Determinants of Consumers’ Acceptance and Usage of Mobile-Commerce Application

Abstract

The growth of mobile commerce marketplaces worldwide has been boosted by modern advances in digital technology. However, Privacy and security are still concern in m-commerce application. Since the previous study has researched the link between security and privacy and purpose to use, the factors that influence the formation of privacy and security in m-commerce are mostly unidentified. On the basis of UTAUT2, this study investigates the factors of security and privacy effecting mobile commerce acceptance. A hybrid SEM-ANN method was utilized to identify non-linear and compensatory interactions. Compensatory and Linear models are based on the idea that a deficiency in one component might also be compensated for by other variables. Linear and Non-compensatory models, on the other hand, seem to overcomplicate buyer decision mechanisms. Survey criteria have been conducted to obtain 890 mobile commerce consumer’s datasets utilizing an application on m-commerce. The following are the results. (1) M-commerce acceptability and use were positively influenced by five determinants (Security, performance expectancy, effort expectancy, habit, and price value). (2) Un-authorization, Error, secondary usage, collection, control, and awareness were all shown to directly and significantly negatively impact M-COMMERCE acceptance and use. (3) Three additional variables (social influence, hedonic motivation, and facilitating conditions) did not affect customers' intentions to use m-commerce applications in Jordan. In m-commerce, the integrated model expects a 45% percent increase in security and privacy.