Sentiment Analysis of COVID-19 Vaccine Tweets Using Machine Learning

Abstract

The SARS-CoV-2 coronavirus disease (COVID-19) pandemic continues to impact the health and well-being of the global population. The pandemic has shifted our view of the world to a different dimension. Control of the increasing spread of COVID-19 lies in vaccinating the public to halt the pandemic outbreak. However, there is uncertainty in the minds of the people regarding vaccines. In this research, we have analyzed the public tweets from Twitter related to COVID-19 vaccinations to detect the user’s view on the vaccination using Machine Learning (ML). Three different ML algorithms viz. Multinomial Na ive Bayes (MNB), Support Vector Machine (SVM), and Logistic Regression (LR) were used for the analysis of the Twitter data into positive or negative tweets. The LR algorithm gave the best results in the analysis with an accuracy of 97.3 %, whereas the accuracy of SVM was 96.26 %. The MNB had the lowest accuracy of 88 % in comparison to the other ML algorithms.