Implementasi LDA untuk Pengelompokan Topik Twitter Bertagar #Mypertamina

Abstract

Abstract  Twitter social media is widely used by users as a medium of communication and information. Apart from being a communication tool, Twitter is used to obtain the required research data. The use of the twitter hashtag becomes a reference for trending news or issues that are developing in the community. The trend that is currently being discussed is the Mypertamina application. This study takes data from twitter with the hashtag #Mypertamina with a lot of twitter data as many as 149 tweets, from the data obtained it will be clustered using topic modeling with the Latent Dirichlet Allocation (LDA) method. The advantage of the LDA method is that it can cluster, summarize, and link large amounts of data. This study resulted in 3 data clusters with the largest coherence value of 0.4618 Keywords: 3-5 keywords; Mypertamina, Twitter, LDA