Análise de estresse e resiliência durante a pandemia da COVID-19 a partir de dados de redes sociais

• Ano de defesa: 2023
• Autor(a) principal: Peres, Diansley Raphael dos Santos
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Ciência da Computação
• Palavras-chave em Português: Estresse; Resiliência; TEPT; TensiStrength; Redes Sociais; Stress; Resilience; PTSD; Social Media; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO; ODS::ODS 3. Saúde e bem-estar - Assegurar uma vida saudável e promover o bem-estar para todos, em todas as idades.
• Link de acesso: https://repositorio.ufu.br/handle/123456789/41249; http://doi.org/10.14393/ufu.di.2023.658

The coronavirus pandemic (SARS-CoV-2) has ravaged countries worldwide since 2019, when it emerged in China. In Brazil alone, by October 2023, more than 37 million disease cases had been confirmed with approximately 706 thousand deaths. In addition to the various physical effects of the COVID-19, psychological effects could be seen through the development of diseases such as alcoholism, depression and post-traumatic stress. It is clear that, during the pandemic, people used social networks for different purposes, so identifying published content can contribute to an adequate response by authorities in emergencies. This work therefore sought to investigate the impact of COVID-19 on perceived stress, resilience and Post-Traumatic Stress Disorder (PTSD) in volunteers from the Federal University of Uberlândia using data from social networks. Self-evaluative questionnaires were also applied to measure the psychological aspects investigated in mining. As one of the research results, it was possible to validate the algorithm for detecting stress and relaxation in texts, TensiStrength, adapted to portuguese. The databases extracted with the volunteers' characteristics made it possible to correlate form data with the stress and relaxation present in the texts and statistics of users' publications on online social networks. Despite the insufficient sample size of the user bases, it was possible to find evidence that X is a network with a greater presence of stress compared to Instagram. In general, the stress present in the text was not directly proportional to the stress perceived in the user. Correlations were verified between users' resilience factors and publication statistics extracted from social networks.