Comparison of Support Vector Machine and XGBSVM in Analyzing Public Opinion on Covid-19 Vaccination

Abstract

The corona virus has become a global pandemic and has spread almost all over the world, including Indonesia. There are many negative impacts caused by the spread of COVID-19 in Indonesia, so the government takes vaccination measures in order to suppress the spread of COVID-19. The public's response to vaccination was quite diverse on Twitter, some were supportive and some were not. The data used in this study came from Twitter which was taken using the drone emprit portal, using the keyword, namely "vaccination". The classification will be carried out using the SVM and hybrid methods between SVM and XGBoost or what is commonly called XGBSVM. The purpose of this study is to provide an overview to the public whether the Covid-19 vaccination actions carried out tend to be positive, neutral or negative opinions. The results of the sentiment evaluation that have been carried out can be seen that SVM has the highest accuracy of 83% with 90:10 data splitting, then XGBSVM produces 79% accuracy with 90:10 data splitting.