Sentiment Identification in COVID-19 Specific Tweets

Abstract

In 2020, our world has been hit by a global pandemic of COVID-19, belonging to the family of Coronavirus. Due to the rapid increase in the infection and the death rate, people have started to develop mixed feelings regarding this situation. Therefore, in this study, our sole focus is to analyze the emotions expressed by people using social media such as Twitter etc. Accumulating and studying the concerning tweets will provide aid to elicitate the real emotions during this hard time. The goal of this study is to present a domain-specific approach to understand sentiments manifested within people around the globe regarding this situation. In order to attain this, corona-specific tweets are acquired from twitter platform. After gathering the tweets, they are labelled and a model is developed which is effective for detecting the actual sentiment behind a tweet related to COVID-19. The substantial assessments are performed in bi-class and multi-class setting over n-gram feature set along with cross-dataset evaluation of different machine learning techniques in order to develop the model. Our experiments reveal that the proposed model performs well in perceiving the perception of people about COVID-19 with a maximum accuracy of about 93%.