ABSTRACT
As people progressively resort to twitter to express their opinions or to disambiguate their sentiment, it's feasible to analyze the mass opinion to conclude the polarity of the subject at hand using sentiment analysis. Sentiment Analysis (SA) has revolutionized the way information is perceived today. Inspired by this, the work in this paper investigates the much-debated act- the Citizenship Amendment Act (CAA) by analyzing opinionated geo-tagged tweets, manually annotated and cross verified by six annotators. This is the first paper to the best of our knowledge to analyse CAA using SA and to provide a clear statistics of the mass opinion across the states of the nation. In this paper, machine learning approach is used for sentiment analysis of tweets. Support vector machine classifier is used to classify the tweets into three classes viz. positive, negative and neutral.
