Twitter Spatio-temporal Topic Dynamics and Sentiment Analysis During the First COVID-19 Lockdown in India

The advent of COVID-19 has tremendously affected the global economy. More people have suffered, and some of them even lost their lives. As the worst part, many countries are seeing second and third waves of COVID-19 cases despite the vaccination. India is one such country that is worse affected by the COVID-19. Understanding what people of India think and how they express their thoughts through social media platforms like Twitter has a vast significance. Therefore, in this project, we aim to use COVID-19-related tweets during the first COVID-19 lockdown in India and apply the nonnegative tensor factorization (NTF) algorithm to elicit spatio-temporal topic dynamics. While knowing the sentiments of people is important, identifying sentiments for each tweet is time-consuming and hard to interpret. Therefore, in this paper, we propose to apply sentiment analysis on the topics identified using NTF.