Opinion mining on COVID-19 vaccines in India using deep and machine learning approaches

Abstract

Since the COVID-19 outbreak, considering the people’s opinion has been perceived as the most crucial challenge for the government to combat the pandemic, such as implementing a national lockdown, instituting a quarantine procedure, providing health services, and more. Furthermore, the government made many critical decisions based on public opinion to combat coronavirus. Opinion mining or sentiment analysis has arisen as a method for mining people’s views on several issues using machine learning techniques. With the support of machine learning methods, this paper extracted the Indian people’s opinions on vaccines through Twitter tweets. More than four lakh vaccine-related tweets from May 04 to May 11, 2021, and from Aug 13 to Aug 21, 2021, were analyzed using state-of-the-art machine learning and deep learning approaches. The BERT and RoBERTa models produced promising results compared to other models on the collected twitter dataset.