ABSTRACT
The Twitter (X) application is very useful for users to get domestic and foreign
news. One of the factors that influences an application is the reviews or ratings given
by users. However, review monitoring can be used to evaluate how users interact with
the Twitter (X) application based on the reviews and ratings they provide. The data
used in this research is a dataset of user reviews of the Twitter (X) application from the
Google Play Store. This research aims to determine the performance of the two
classification algorithms of the Support Vector Machine and Naive Bayes methods to
be implemented through a classification model in the performance evaluation training
set by calculating the accuracy that has been obtained from the Support Vector
Machine classification, getting an accuracy result of 80%, while the accuracy obtained
by Naive Bayes 78% The results of this comparison are expected to provide insight into
the effectiveness of both methods in classifying the sentiment of user reviews of the
Twitter (X) application.
Keywords: Sentiment analysis, Naïve Bayes, Support Vector Machine, Twitter(X).