Open AccessTwitter Sentiment Analysis of Bangkok Tourism During COVID-19 Pandemic Using Support Vector Machine Algorithm
Author(s) -
Thanapat Sontayasara,
Sirawit Jariyapongpaiboon,
Ar Promjun,
Napat Seelpipat,
Kumpol Saengtabtim,
Jing Tang,
Natt Leelawat
Publication year - 2021
Publication title -
journal of disaster research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.332
H-Index - 18
eISSN - 1883-8030
pISSN - 1881-2473
DOI - 10.20965/jdr.2021.p0024
Subject(s) - tourism , sentiment analysis , pandemic , social media , destinations , covid-19 , china , business , geography , disease , artificial intelligence , computer science , medicine , world wide web , archaeology , pathology , infectious disease (medical specialty)
In the year 2020, SARS-CoV-2, the virus behind the coronavirus disease (COVID-19) pandemic, affected many lives and businesses worldwide. COVID-19, which originated in Wuhan City, China, at the end of December 2019, spread over the entire world in approximately four months. By October 2020, approximately 20 million people were infected and millions had died from this disease. Many health organizations such as the World Health Organization and Centers for Disease Control and Prevention made COVID-19 their primary focus. Many industries, especially, the tourism industry, were affected by the pandemic as many flight and hotel reservations were canceled. Thailand, a country considered one of the world’s most popular tourist destinations, suffered much losses because of this pandemic. Many events and travel bookings were canceled and/or postponed. Many people expressed their views and emotions related to this situation over social media, which is considered a powerful media for spreading news and information. In this research, the views of people who were planning to travel to Bangkok, the capital city and most popular destination in Thailand, were retrieved from Twitter for the dates between April 3 and 30, 2020, the period during which the country underwent nationwide lockdown. Sentiment analysis was performed using the support vector machine algorithm. The results showed 71.03% classification accuracy based on three sentiment classifications: positive, negative, and neutral. This study could thus provide an insight into travelers’ opinions and sentiments related to the tourism business. Based on the significant terms in each sentiment extracted, strengths and weaknesses of each tourism issue could be obtained, which could be used for making recommendations to the related tourism organizations.