Assessment and Modeling of Decision-Making Process for e-Commerce Trust Based on Machine Learning Algorithms

Decision-making on trust in an e-commerce environment is associated with several other elements such as security, risk, satisfaction, loyalty and reputation. The life cycle of e-trust in online Business to Consumer (B2C) transactions takes multiple stages from beginning to end. The first stage involves the complete unawareness of online shoppers about online sellers. Then, individual trust begins to develop, endure, recover and ultimately sustain or diminish. Since it is a complicated task to gain trust, there have been numerous solutions, methods and models proposed so far to create, maintain, measure, enhance and prevent loss of trust. One of the solutions that has long been adopted to determine trust level in B2C is a concentration on the history or background of online shoppers (customers) and online sellers (companies) so as to obtain reliable data or identify trust level. This paper attempted to adopt the machine learning algorithms to analyze decisions about the past and history of individuals/companies and trust in e-business/e-commerce (EC). Moreover, efforts were made to identify and assess the key contributing factors to decision-making. The results demonstrated that corporate factors and business models left the greatest impacts on customer decision-making in e-business trust or distrust during electronic transactions.