Sentiment Analysis on Multimodal Transportation during the COVID-19 Using Social Media Data

This paper aims to leverage Twitter data understand travel mode choices during the pandemic. Tweets related different modes in New York City (NYC) are fetched from two most recent years (January 2020–January 2022). Building on these data, we develop classifiers, adapted natural language processing (NLP) models, determine whether individual tweets some (subway, bus, bike, taxi/Uber, and private vehicle). Sentiment analysis is performed people’s attitudinal changes about Results show that a majority of people had positive attitude toward buses, bikes, vehicles, which consistent with phenomenon many commuters shifting away subways bikes vehicles We analyze negative find were worried those who did not wear masks buses. Based users’ demographic information, conduct regression what factors affected public transit. users service industry was more easily by MTA subway