Análise de sentimentos sobre as vacinas contra Covid-19: um estudo com algoritmo de machine learning em postagens no twitterr

Abstract

Introduction and objective: Covid-19 vaccines were among the most talked-about topics during the pandemic on the social network Twitter. In this paper a Sentiment Analysis was performed on Twitter users' posts. Methods: The Python 3 programming language was used, applying specific libraries to collect, manipulate and visualize the tweets and the data. A mapping of the sentiments of Brazilian users was performed for the period January 17 to March 6, 2021, classifying them into positive, neutral, and negative. Results: A significant variation in sentiment polarity was perceived, with a reduction in positive sentiments and consequent growth in negative and neutral ones during the period. This type of analysis acquires research relevance when one considers a large amount of information and the speed with which it is shared on the networks. The results confirm that the tweets referring to Covid-19 influenced the users' sentiment. Conclusion: The feeling that information produces in people can directly or indirectly impact their lives and society in general, presenting significant relevance, since sentiment can be an important factor in the user experience.