Analysis Report for Statistics in the Twitter Network

Abstract

Twitter is the most popular social networking service across the world. In Twitter, the messages are known as tweets. Tweets are mainly text-based posts that can be up to 140 characters long which can reach the author's subscribers. These subscribers are also known as followers. Such subscriptions form a direct connection. But these connections are not always symmetric. In this study, the authors have assumed that if two nodes are connected, then the tweet is propagated between them without any other conditions. But using sentiment analysis, the general opinion of people about various things can be figured. The Twitter data set analyzed includes almost 20k nodes and 33k edges, where the visualization is done with software called Gephi. Later a deep dive analysis is done by calculating some of the metrics such as degree centrality and closeness centrality for the obtained Twitter network. Using this analysis, it is easy to find the influencers in the Twitter network and also the various groups involved in the network.