An analysis of ridesharing trip time pre- and amid-COVID-19 pandemic using advanced text mining technique – the USA vs Indian case study across different age and gender groups

Abstract

Abstract With the spread of the Covid-19 virus, the public transportation industry faced new challenges. This disease may have affected the decision of users in choosing travel modes by diluting the strengths of ridesharing. In this study, our aim was to investigate the opinions of users on the Twitter application pre- and post-pandemic about travel time in ridesharing. For this purpose, we analyzed 63,800 tweets from January 1, 2019 to April 30, 2022, focusing on the countries of the United States and India, taking into account the characteristics of users such as age and gender. The method we used was LDA for topic modelling and BERT for sentiment analysis. Our results show that trip time happening, such as mornings and other times, became more important after COVID-19 and users' concern about waiting time has increased. After the pandemic, men are more worried about the waiting time, while women and younger groups are worried about the time cost. Before the pandemic, older people gave more importance to waiting time, but after the pandemic, they are more concerned about the time cost and the trip time happening. We also found that women and US users were more positive than others about ridesharing trip time in the pre-pandemic era. Our findings showed that two main factors, the pandemic and the country, greatly affect sentiment, and the positive sentiment of the users of both countries increased after the epidemic, which was higher among all groups for the elderly and American users. Our results can be used to analyze the sentiments of ridesharing industry users in order to compete with other public transportation companies by providing better services and designing a standard to improve travel time.