Twitter Sentiment Analysis for COVID-19

With the current onset of the Coronavirus pandemic social interaction has been deeply affected. However, social media and micro-blogging platforms like Twitter are becoming more of relevance to the masses to express their feelings, opinions, concerns and problems. A large amount of data is generated in this process which may be highly valuable in deriving deep insights about the effects of the pandemic on the people. With modern technological tools like data mining, data processing and APIs, harnessing this sea of data has become more of a possibility. Moreover, with the application of advanced machine learning tools like sentiment analysis, understanding the people's mindset behind the tweet has also become a possibility. The study oversees the development of a web based interface for connecting two sections of the society: one which is in need of help and the other with the desire to help. In this study, we aim to utilise data analysis and advanced machine learning techniques like RoBERTa(Robustly Optimised BERT pre-training Approach) and CNN-RoBERTa Sentiment Extraction to incorporate methods like sentiment analysis, frequency distribution and comparative analysis on data found from social media for efficient comparison on the effects of COVID-19 in India while also developing a web-based interface for effective exposition and insights over the crisis.