Analysis and Mining of Sentiments in Public Health and Social Measures Related Corpus

Sentiment Analysis is a rich research topic, and since the COVID-19 outbreak there has been a high rate of research on identifying the different sentiment of opinions expressed online. Government declarations, Vaccination drives, Crypto-currency prices, etc. are some of the many contexts that have gathered user attention during the pandemic, and consequently the relevant social media posts have been researched. In most of the state-of-the-art works, it is noted that the data is acquired from online social media platforms such as Twitter. However, in this work, we have utilized a dataset provided online by the World Health Organization (WHO) based on the different public health and social measures taken by the local authorities around the globe in response to the spread of COVID-19. On analyzing the different texts pertaining to the enlisted events, we have identified that about 40% of the different posts and articles that have been presented in this dataset convey a negative sense. It is also observed that highest negativity of 46.4% exists in the articles related to the African Region (AFR), and the maximum positivity of more than 30% is observed in the articles related to the American Region (AMR).