Leveraging Social Media to Understand Public Perceptions toward Micromobility Policies: The Dallas Scooter Ban Case

Abstract

In this study, comments made on social media by City of Dallas residents were analyzed following a citywide scooter ban to reveal hidden attitudes towards micromobility. Using text mining techniques, results showed that the majority of residents were in fact against the ban. By applying term frequency-inverse document frequency (tf-idf), key terms mentioned by ban supporters and opponents were uncovered. Latent Dirichlet Allocation further showed that topics such as Bike, Public Safety, Authorities, Cars, and Feelings were discussed most frequently within resident comments. Findings demonstrate the potential for social media to clarify public perceptions towards public services such as micromobility.