Public Opinion on Dockless Bike Sharing: A Machine Learning Approach

Abstract

Dockless bike sharing is an emerging paradigm. Like many other technologies, it brings advantages and disadvantages to communities. Further investigation into public opinion will shed light on the impact of this technology on communities and provide input to city authorities for transportation planning. Transportation planning processes can be enhanced by engaging the community through social media technologies. Social media like Twitter, Facebook, and other microblogging media have been used for planning, but have not been extensively evaluated for that purpose. This study examined approximately 32,000 posts on Twitter to assess public opinion on dockless bike-sharing systems. Using a mix of text mining and statistical techniques, we examined relevant posts to determine the sentiment polarity of tweets, the underlying topics in the tweets, and the extent of engagement and impact on the decision-making process. Results given by two different sentiment algorithms show that there is more positive than negative polarity across the algorithms. Also, the findings show that the underlying topics in tweets include electric scooters, private e-hailing companies, and blockage of sidewalks, among others. The results indicate that the dockless shared mobility models are potentially useful in generating participation, but faced substantial technical, analytical, and communication barriers to influencing decision-making.

How to Save Bike-Sharing: An Evidence-Based Survival Toolkit for Policy-Makers and Mobility Providers

“…This study also reported that peer-to-peer sharing is likely to remain a niche with special purpose bikes. In dock-based schemes, satisfaction is flawed by a set of factors such as the mechanics of the bikes, the picking and dropping system, and the apps used to organise the service [34], while dockless shared mobility models are potentially useful in generating participation but face substantial technical, analytical, and communication barriers [35]. The rapid expansion of dockless bike-sharing may be the reason behind a large-scale renaissance of the very concept since 2016, however, this is coincident with the serious oversupply of bikes [36] in many programmes.…”

Monitoring the first dockless bike sharing system in Greece: Understanding user perceptions, usage patterns and adoption barriers

“…Furthermore, in dockbased schemes satisfaction is seriously affected by a set of factors like the mechanics of the bikes, the picking and dropping system, and the apps used to organise the service (Manzi and Saiben, 2018). At the same time dockless shared mobility models are potentially useful in generating participation, but face substantial technical, analytical, and communication barriers (Taleqani, et al, 2019) and higher vulnerability to mismanagement (Luo et al, 2019).…”

Machine learning for geographically differentiated climate change mitigation in urban areas

“…For example, natural language processing tools can identify geo-localized urban activities from publicly available text obtained from, e.g. Twitter (Rahim Taleqani, Hough, & Nygard, 2019), where location information can be extracted and associated to activities of relevance.…”