Analisis Sentimen Pada Media Sosial Twitter Mengenai Tanggapan Vaksinasi COVID-19 Menggunakan Metode Naive Bayes

Abstract

Coronavirus has become a pandemic in the world and has spread to almost all parts of the world, including in Indonesia. Many negative impacts resulted from the spread of covid-19 in Indonesia, so the government took action to implement vaccination programs to reduce the spread of covid-19. The response from the public to the vaccination program is quite diverse on social media Twitter, some agree and some disagree. this type of comment on twitter in the form of unstructured text in large numbers resulted in the missed information about COVID-19 vaccination that is useful from a set of text documents. It also knows the sentiment of twitter users manually can be detrimental to time and effort. This study aims to determine public sentiment towards vaccination. The data used are 1000 tweets, using two keywords, namely "vaccine + covid". The data is then divided into training data and test data. Classification was carried out using the Naïve Bayes method and using 10-Fold Cross Validation. The classification results from the Naïve Bayes method get an average accuracy of 73.75%.