Analysis of Social Trends for Electric Scooters Using Dynamic Topic Modeling and Sentiment Analysis


KIPS Transactions on Software and Data Engineering, Vol. 12, No. 1, pp. 19-30, Jan. 2023
https://doi.org/10.3745/KTSDE.2023.12.1.19,   PDF Download:
Keywords: Personal Mobility Vehicle, Electric scooter, Text Mining, Dynamic Topic Modeling, sentiment analysis
Abstract

An electric scooter(e-scooter), one popularized micro-mobility vehicle has shown rapidly increasing use in many cities. In South Korea, the use of e-scooters has greatly increased, as some companies have launched e-scooter sharing services in a few large cities, starting with Seoul in 2018. However, the use of e-scooters is still controversial because of issues such as parking and safety. Since the perception toward the means of transportation affects the mode choice, it is necessary to track the trends for electric scooters to make the use of e-scooters more active. Hence, this study aimed to analyze the trends related to e-scooters. For this purpose, we analyzed news articles related to e-scooters published from 2014 to 2020 using dynamic topic modeling to extract issues and sentiment analysis to investigate how the degree of positive and negative opinions in news articles had changed. As a result of topic modeling, it was possible to extract three different topics related to micro-mobility technologies, shared e-scooter services, and regulations for micro-mobility, and the proportion of the topic for regulations for micro-mobility increased as shared e-scooter services increased in recent years. In addition, the top positive words included quick, enjoyable, and easy, whereas the top negative words included threat, complaint, and ilegal, which implies that people satisfied with the convenience of e-scooter or e-scooter sharing services, but safety and parking issues should be addressed for micro-mobility services to become more active. In conclusion, this study was able to understand how issues and social trends related to e-scooters have changed, and to determine the issues that need to be addressed. Moreover, it is expected that the research framework using dynamic topic modeling and sentiment analysis will be helpful in determining social trends on various areas.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
K. Kim and Y. Shin, "Analysis of Social Trends for Electric Scooters Using Dynamic Topic Modeling and Sentiment Analysis," KIPS Transactions on Software and Data Engineering, vol. 12, no. 1, pp. 19-30, 2023. DOI: https://doi.org/10.3745/KTSDE.2023.12.1.19.

[ACM Style]
Kyoungok Kim and Yerang Shin. 2023. Analysis of Social Trends for Electric Scooters Using Dynamic Topic Modeling and Sentiment Analysis. KIPS Transactions on Software and Data Engineering, 12, 1, (2023), 19-30. DOI: https://doi.org/10.3745/KTSDE.2023.12.1.19.