City Transit Rider Tweets: Understanding Sentiments and Politeness

Document Type

Journal Article

Publication Date


Subject Area

place - north america, place - urban, technology - passenger information, ridership - perceptions, planning - service improvement


Social media, Twitter, transit, sentiment analysis, politeness measure


With the expanding popularity of Web 2.0, there has been a huge surge in the use of social media, like Twitter, to express user sentiments or opinions. Delays and breakdowns in transit operations can make riders annoyed and irritated, and as a result, they express their anger and frustration via social media posts. Understanding the tipping points of public frustration will help in developing better solutions. This study aims to develop a framework by developing multilevel sentiment analysis and determine the emotion and politeness measures using transit-related tweets from New York (New York City) and California (San Francisco). The popular hashtags associated with the transit systems of New York and California were collected during 2019. The words associated with negative sentiments widely differ in these two states. Moderate levels of differences are seen in the politeness measures for these two states. Additionally, co-occurrence measures associated with negative emotions identified unique issues based on the demographics. This study demonstrates that Twitter provides a great opportunity to understand the public perception of transit, and the findings can help authorities design a more efficient transit system to improve user experience.


Permission to publish the abstract has been given by Taylor&Francis, copyright remains with them.