Analyzing COVID-19 sentiments on Twitter: A machine learning approach
V. Aishwarya, K. Sanmuga Priyan, R Vyshnavi, M. Azhagiri, S. Sathya Priya · Computational Methods in Science and Technology · 2024
Throughout the outbreak, COVID-19 has been a hot issue on Twitter, providing a forum for people to voice their ideas and perspectives. Unfortunately, there isn&s;t a thorough rundown of sentiment analysis in Twitter discussions around COVID-19. By analyzing sentiment analysis utilizing COVID-19 Twitter data from both a machine learning and a behavioral and social science standpoint, this article fills this gap. Forty papers examining public attitudes were found in a study of the literature conducted between October 2019 and January 2022. Higher accuracy is shown by ensemble models, especially those that use BERT and RoBERTa. These models do not provide exact outcomes, despite their theoretical promise. Based on the COVIDSENTI dataset, the research emphasizes the need for government action on social media to counter false information. It highlights that, despite the continual examination of datasets, proactive public health actions are crucial in lowering unfavorable sentiments during a pandemic.