An Investigation into the Impact and Predictability of Emotional Polarity on the Virality of Twitter Tweets

Abstract

In this study, the likelihood of tweet-sentiment influencing people's opinion and motivating people to re-tweet is examined using data pertaining to three different fields; political, entertainment and financial. This study carried out additional investigation into the impact of Sentiment Compound Polarity and Favourite Count on people's opinions. Metrics of retweets and users' favourite count are used for the development of a predictive model to determine the likelihood of an individual tweet being retweeted. Public datasets were used for this study, focusing on three diverse topics including the 2017 demonetization in India dataset with 14,940 observations, the 2016 US election dataset with 397,629 observations and the 2018 American Music Awards dataset with 27,556 observations. Findings demonstrate that tweet sentiment plays an important role in shaping people's views and thereby inspiring them to retweet a tweet. The resulting predictive model may be used to determine the likelihood of a tweet being retweeted. The outcomes have numerous applications in domains such as advertising and marketing, political, social, commercial and charitable organisations.