Analysis of Public Opinion on Covid-19 Vaccine through Social Media Using Naïve Bayes Theory Algorithm

Abstract

This study aims to analyze various public opinion on the Covid-19 vaccines that appear on social media pages, especially on Facebook and Twitter via # (hastag). The death rate caused by COVID-19 was so high which reached 144,227 people until 2022. The Indonesian government required vaccines for the community starting from children aged 6 years as an effort to prevent the spread of the Covid-19 virus. Unfortunately, the implementation of complete vaccines in Indonesia has only reached 51.3% of the mandatory vaccine population, which is 140 million out of 339 million people. The non-achievement of the target set by the government causes the need to conduct a sentiment analysis on vaccines in Indonesia through social media. Based on the sample data, from 1000 words obtained from 320 opinions there are positive and negative opinions. This data is then analyzed and processed to find out how many positive and negative responses occurred. The data was then processed into several stages to test the level of truth through training data and test data. The results of the data processing were tested using the Naïve Bayes algorithm which resulted in an accuracy value with a precession of 77.08% taken from 90 samples test data, recall with a percentage of 97.87% based on positive data which was predicted to be true with a positive opinion status from 47 samples of test data and 1 positive data status which is still predicted to be negative. Furthermore, the specific percentage value obtained was 65.30% of the 132 test data that are predicted to be true negative.