Acceptance as situated judgement: text mining user reviews of autonomous public transport

Understanding how shared autonomous vehicles shape travel behaviour, accessibility, and urban form is essential for ensuring that their diffusion contributes meaningfully to sustainable accessibility. However, as adoption remains in its early stages, prior research has relied largely on surveys of potential users that capture hypothetical scenarios and abstract expectations. This study addresses this gap by examining how actual users of autonomous public transport (APT) describe and interpret their experiences, drawing on large-scale reviews. Applying text mining, topic modelling, and network analysis, this study identifies the main themes in their reviews and explore how perceptions vary across spatial, temporal, and operational contexts. The findings reveal that APT acceptance is best understood as a ‘situated judgement’ shaped by a combination of functional, technological, and emotional factors, as well as local contexts. These insights emphasise that successful APT deployment requires more than just technical reliability – sensitivity to user experience and alignment with local urban form, land-use context, and the meanings attached to destinations.