Predicting the Determinants of Augmented Reality Technologies in E-commerce Applications: A SEM and ANN Modeling

Abstract

Augmented reality (AR) technologies are the most promising technologies for e-commerce applications. This study has used the UTAUT theory, extending with privacy risk, which may impact the user’s decision to try AR features in e-commerce apps. The paper investigated factors using a two-stage approach comprising of Structural Equation Model (SEM) and an Artificial Neural Network (ANN). The validation of the model is performed with the IBM SPSS AMOS 24 and IBM SPSS 25 software. The results revealed performance expectancy as a noteworthy determinant through the analysis. Further privacy risks involved while using AR features of online shopping apps negatively impact the influence on the choice of the users to try out the technology. The proposed model has explained the user behavioral intention with 36.7%, whereas using ANN has predicted with an accuracy of 48.9%, indicating the model has high predictive power. This study’s theoretical and practical insights will contribute to developing and refining augmented reality systems in e-commerce applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.