Twitter data sentiment analysis of tourism in Thailand during the COVID-19 pandemic using machine learning

Natt Leelawat; Sirawit Jariyapongpaiboon; Arnon Promjun; Samit Boonyarak; Kumpol Saengtabtim; Ampan Laosunthara; Alfan Kurnia Yudha; Jing Tang

doi:10.1016/j.heliyon.2022.e10894

Twitter data sentiment analysis of tourism in Thailand during the COVID-19 pandemic using machine learning

Heliyon. 2022 Oct;8(10):e10894. doi: 10.1016/j.heliyon.2022.e10894. Epub 2022 Oct 3.

Authors

Natt Leelawat^{1

2}, Sirawit Jariyapongpaiboon¹, Arnon Promjun¹, Samit Boonyarak³, Kumpol Saengtabtim¹, Ampan Laosunthara², Alfan Kurnia Yudha¹, Jing Tang^{2

3}

Affiliations

¹ Department of Industrial Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand.
² Disaster and Risk Management Information Systems Research Unit, Chulalongkorn University, Bangkok, Thailand.
³ International School of Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok, Thailand.

Abstract

The coronavirus disease 2019 (COVID-19) pandemic has severely affected Thailand's economy, which relies heavily on tourism. In this study, we labeled the sentiment and intention classes of English-language tweets related to tourism in Bangkok, Chiang Mai, and Phuket. Then, the accuracy of three machine learning algorithms (decision tree, random forest, and support vector machine) in predicting the sentiments and intentions of the tweets was investigated. The support vector machine algorithm provided the best results for sentiment analysis, with a maximum accuracy of 77.4%. In the intention analysis, the random forest algorithm achieved an accuracy of 95.4%. In a subsequent preliminary qualitative content analysis, the top 10 words found in each sentiment and intention class were gathered to provide insights and suggestions to help increase tourism in Thailand. The results of this study suggest that to help restore tourism in Thailand, tourist destinations, natural attractions, restaurants, and nightlife should be promoted. In addition, the two main concerns of tourists to Thailand should be addressed: COVID-19 and current political tensions.

Keywords: COVID19; Machine learning; Sentiment analysis; Thailand; Tourism; Tweet.