This study aims to examine the ways in which data-driven personalised fashion systems influence users' decision-making processes via a set of contextual variables.
A quantitative approach using a structured questionnaire was adopted to investigate how technical features of data-driven personalised fashion recommendations (i.e. accuracy, relevance, interactivity, diversity, security, emotional connection and social influence) affect user satisfaction and purchase intention through perceived usefulness and trust. 534 survey results were examined using the structural equation modelling (SEM) technique.
The findings show that user perceptions of decision satisfaction, perceived usefulness, perceived trust and purchase intention are significantly impacted by the technological features of personalised recommendation systems. Enhancing user pleasure and buying decisions will ultimately result from increased user trust and perceived usefulness.
These results offer useful theoretical justification and real-world applications for improving the architecture and deployment of customised recommendation systems.
