Social network analysis and visualization for general election in Malaysia

Abstract

Social Media's presence is ubiquitous today. Obviously, we can see that internet and information has grown up to take part in people's lives. Internet platform is expected to play a bigger important role in politics, and this can increase the skills of politicians in seeking for vote. Thus, the main issue is about how this social media can be a factor that will transform the voting results on the networking site. Therefore, Social Network Analysis is proposed for the study. The purpose of the study is to identify the related variable for among different political parties, to develop and visualize social network graph based on graph theory concept and to predict the most voter based on the party leader using classifier method Logistic, Naive Bayes and J48. The dataset for the study has been crawling from social media such as Twitter and Facebook, starting from July 2017 until October 2017. Based on the outcome result, it can be concluded that the influence of candidates in a party has a great impact on seeking voters. This is proven through research made by observing the influence of BN party leader, Najib Razak, whose number of followers are high and thus winning the major voting. The margin of votes are significant compared to the other party leader such as Lim Kit Siang, DAP candidate, despite the frequent updates on the official website. Hence the number of follower that influencing voting made.