Sentiment Analysis of Twitter Data Related to COVID-19

Abstract

Sentiment analysis is natural language processing technique which is used to evaluate the emotions of subjective data. Machine learning along with natural language processing is used to evaluate weighted values of sentences or phrases. The COVID-19 pandemic is one of the greatest challenges that humans have faced since World War II. With the first case of COVID-19 being reported there has been an urge of discovering efficient analytic methods to understand mass sentiment in the pandemic scenario. Along with this technique, various strategies are used such as computational etymology and information retrieval. The task of detecting text polarity becomes easier with sentiment analysis and we can classify like “negative”, “neutral”, “positive”, or “impartial”. The main objective is using or studying the methods of sentiment analysis of Twitter data and provide comparative study of mass sentiment during the chosen window, i.e., January to August. This chapter has explored Twitter as dataset. TextBlob which is a sentiment analysis library in Python was used to extract the objective of the content and also to know whether the content is positive or negative.