Open AccessAnalyzing Netizens’ Perceptions Towards Indonesian Presidential Candidates Using Topic Modeling Approach
Author(s) -
Devi Karolita,
Ariesta Lestari
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/546/5/052037
Subject(s) - microblogging , social media , topic model , latent dirichlet allocation , event (particle physics) , indonesian , perception , presidential system , presidential election , computer science , world wide web , data science , psychology , political science , politics , information retrieval , linguistics , philosophy , physics , quantum mechanics , neuroscience , law
Over the past few years, Twitter has significantly grown as the microblogging platform. Millions of user use this platform to share their attitudes, views, and opinion on a daily basis. This phenomenon has been used to promote people’s attention towards some event, such as 2019 Indonesian Presidential Election. In this study, we investigate people’s online opinions towards the event through social media. The goal of the study is to discover frequent topics amongst netizens’ tweets during the election campaign. We collected tweets containing the names of the candidates, then applied topic modelling approach using Latent Dirichlet Allocation (LDA) method to cluster the topics. Based on the experiment, the tweets are clustered into ten topics with different focuses e.g., a topic discusses the candidate’s position towards sensitive issues, a topic about the community supports towards one presidential candidate. Our result shows that topic modelling approach can be used to analyse people’s perception in social media towards an important event.