Machine learning-based empirical investigation of user’s perception of digitalisation in pandemic immunisation programs

Abstract

ABSTRACT This study proposes and validates a framework for tracking post-usage perceptions of digitalisation in pandemic immunisation programmes (PIPs) using machine learning techniques. We categorise PIP digitalisation features into two groups: those offering benefits and those requiring effort/cost. Using consumer value theory (CVT), we examine post-usage perceptions of these features. Online review data from developed and developing economies are analysed, with topic modelling applied to over 46,000 reviews to identify key PIP digitalisation features in two nations. An empirical investigation using dominance and correspondence analysis was conducted to understand their influence on users’ perceptions. The regression coefficient sign indicates the positive or negative perception associated with a feature. This framework helps identify poorly perceived digitalisation features and guides decisions on remedial actions to improve post-usage perception. By drawing on this theoretical framework, we gain insights into how users perceive the value of PIP digitalisation features.