ABSTRACT
The Covid-19 pandemic has severely affected many countries around the globe in terms of physically as well as mentally. During the initial months of the pandemic have reported India’s deficient cases, but eventually the cases were proliferated as the time progress. The government’s decision to impose a lockdown without warning has a wide-ranging impact, affecting everyone from low-wage workers to huge corporations. As a result, there is a negative impact on people’s mental health and emotions. The people had suffered from depressions, anxiety, fatigue and so forth. Many wide varieties of the people had expressed their thoughts, viewpoints, and their mental conditions in the form of tweets over the Twitter, a social media platform. Hence, in this paper, we have statistically analysed the data of tweeted tweets to elicit the meaningful insights. The data was analysed using the unsupervised clustering strategy–K-means and LDA–and the results were reinforced and validated using the pre-trained supervised classification approach–Text to Text transformer. The anticipated data depicted that the fear was the most common state of mind at the end of the lockdown, followed by joy, anger, and sadness. Furthermore, the deduced insights will be highly beneficial in decision-making process when such an epidemic or pandemic situation re-surges. © 2021, Springer Nature Switzerland AG.