Fine-Tuning BERT-Base-Cased Model on COVID-19 Vaccine Tweets

Massive volumes of data are available on the World Wide Web in the form of user thoughts, feelings, opinions, and discussions about various social events, products, brands, and political concerns through sites like social networks, forums, review websites, and blogs. During the COVID-19, Twitter has been a useful source of news and a public platform for discussion. The COVID-19 pandemic offers a chance to do such research that helps to analyse the views of people regarding COVID-19 vaccine. In this paper, a pre-trained unsupervised language model called BERT is used for analysing the views of people regarding COVID-19 vaccine. The tweets were collected using the SNS scrape and sentiment analysis has been done on COVID-19 vaccine tweets using the BERT-base-cased model and achieve the promising results.