Analysis of Twitter Sentiments About the Russian-Ukraine War Using Naive Bayes Based on Particle Swarm Optimization

Abstract

The Russia-Ukraine crisis has not found a solution until now, at least until November 2022. Many Indonesians have expressed their opinions on this matter via social media Twitter. This study uses the Naïve Bayes (NB) Algorithm based on Particle Swarm Optimization (PSO) to analyze opinions regarding the RussiaUkraine war. Data was taken from Twitter using the keywords "Russia Ukraine", "Russia vs Ukraine", "Russia-Ukraine", and "Perang Russia". The number of data sets taken is 5000 data. The results showed that the accuracy of the Naïve Bayes (NB) algorithm without Particle Swarm Optimization (PSO) was 67.72%, the precision value was 58.33%, the recall value was 79.75%, and the error rate of 57.14% while Naïve Bayes (NB) with Particle Swarm Optimization (PSO) the accuracy obtained is 73.48%, the precision value is 65.62%, the recall value is 76.36% and error rate of 50.36%. Thus it can be said that the Particle Swarm Optimization (PSO) applied to the Naïve Bayes algorithm for objects in research can increase the accuracy of results.