Analisis Komentar Toxic Terhadap Informasi COVID-19 pada YouTube Kementerian Kesehatan Menggunakan Metode Naïve Bayes Classifier

Abstract

Countries around the world were shocked by the outbreak of a new virus in 2020, which quickly transmitted and attacks humans of all ages. The virus is COVID-19. The government has advised through social media to stay at home and got vaccinated. YouTube has become a platform for the government, especially the Ministry of Health, to share public information in the COVID-19’s pandemic. Public can put their comments on video uploaded by the Ministry of Health. An analysis of comments is needed so that the information in comments can be useful for those who read and evaluated by the government so that they can provide information that the public can understand. In analyzing toxic comments, it can used text mining. And one of that is the Naïve Bayes Classifier. This study uses the Naïve Bayes Classifier method to determine the results of the analysis. Measuring the value of accuracy, this study using the Confusion Matrix evaluation. From the final result, the highest accuracy value is in the comparison of 90%:10% with an accuracy value of 80%. And from the results of the analysis, the most toxic words used are the words ‘dead’, 'business', ‘public' and 'fool'. From the results show that there are still many people who do not believe in the existence of COVID-19 and think that vaccines can cause death in people who are vaccinated.