Analysis of COVID-19 Information Based on Social Media Big Data Classification Using the K-Means Data Mining Method

Abstract

This Covid-19 began to infect almost all countries in early 2020, including in Indonesia, Covid-19 spread widely throughout the world and was declared as a global pandemic by the World Health Organization (WHO). In the current era of Big Data, large amounts of data have been generated and collected from a variety of rich data sources. Big Data is useful information and valuable knowledge. In this study, the method that will be used for data analysis is the K-Means algorithm with orange tools as a tool to display the results of data classification. One of the information that can be generated is Sentiment Analysis. The purpose of this research was to determining the information such as public sentiment on social media towards government policies in handling COVID-19. In this research, 2000 tweets were used. The keyword used related to government policies are sourced from several online media. The tools used to analyze this twitter data is using Orange Application. The selected keywords are covid19, lockdown, PSBB, and isolation. This keyword is used as a reference to retrieve tweet data from twitter. From each of these keywords, a sentiment classification process will be carried out automatically so that data or tweets are obtained and grouped into positive, negative and neutral sentiment classes. From the result of research conducted, public sentiment on social media towards government policies in handling this virus outbreak tends to be positive.