Machine learning and sentiment analysis to assess the evolution of the COVID-19 pandemic and the impacts on tourism

• Autor(a) principal: Oliveira, Paulo Sergio Gonçalves de
• Data de Publicação: 2023
• Outros Autores: Wada, Elizabeth Kyoko, Lopes, Anderson Soares, Silva, Luciano Ferreira da
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Anais Brasileiros de Estudos Turísticos
• Texto Completo: https://periodicos.ufjf.br/index.php/abet/article/view/39150
• topic: Artificial Intelligence; Hospitality in Tourism; Sentiment Analysis; Hostility in Tourism; Intelligence Artificielle; Hospitalité dans le Tourisme; Hostilité dans le Tourisme; Analyse de Sentiments; Inteligência Artificial; Hospitalidade no Turismo; Hostilidade no Turismo; Análise de Sentimentos

This article aims to analyze how the emotional, mental, and sentimental demands related with hospitality and hostility were developed during the pandemic of the COVID-19 in Brazil. As methological procedures was applied sequential mixed methods research. Firstly, about 1,000 pieces of news were collected from two Brazilian websites to be able to manually classify them in the feelings of hospitality and hostility. We use a machine learning supervisor analysis following a sentiment analysis technique. Secondly, the data were used for training in eight machine learning algorithms, through supervised analysis, being chosen the logistic regression for the data classification, because it fits better to the data, reaching 72% of accuracy. The data collected in two years of the pandemic, thus approximately 221,000 news were then classified using the chosen algorithm, which allowed the generation of graphics and analysis through inferential statistics, through the evolution of feelings of hospitality and hostility. The results indicate that in situations such as the COVID-19 pandemic, people tend to behave hostilely, which leads to a lack of hospitality. The implications of this study are related to the ability to materialize, through the concepts of hospitality and hostility, the perception of visitors, guests, among other people, involved in the tourism sector. Therefore, the sentiment analysis from social media and news affected the tourism and hospitality industry.