Abstract:
Sentiment analysis is a process that aims to determine positive or negative polarity.
Twitter, as a social media platform, is used as a space for public information and
opinions in responding to issues, including the increase in fuel prices. The government
officially increased fuel prices, which was met with criticism from the public. This study
aims to categorize the sentiment polarities of the public and determine the accuracy
level in sentiment analysis. The method used in this study is Support Vector Machine
with linear, RBF, polynomial, and sigmoid kernels, with the linear kernel being
obtained as the best kernel. Based on the results of the testing done on sentiment data
related to the increase in fuel prices in Indonesia on Twitter, which consists of 1130
data (485 positive and 645 negative), evaluation was done using a confusion matrix to
see how well the model can classify correctly. The results show that the Support Vector
Machine method produces an accuracy rate of 88.49%.