SENTIMENT ANALYSIS OF 2024 PRESIDENTIAL CANDIDATE USING THE SUPPORT VECTOR MACHINE ALGORITHM ON TWITTER

Abstract

Indonesian is a democratic country with a huge population and the largest Twitter users in the world. The 2024 presidential election in Indonesia is an interesting topic for Twitter users. Public tweets related to presidential candidates can be used to see a picture of public opinion on presidential candidates. The large number of incoming tweets about presidential candidates encourages the need for methods that help to see public opinion effectively. One method that can be used to classify public opinion effectively is Support Vector Machine (SVM). This method will classify whether a public opinion belongs to a positive or negative sentiment by finding the best hyperlane from both classification classes. The addition of the Kernel function to the Support Vector Machine is useful for dealing with data that is not linearly separated. The use of the K-Fold Cross Validation Method is intended so that data can alternately become test data so as to increase accuracy. Weighting is done using the Term Frequency Document Inverse Frequency (TF-IDF). System evaluation was carried out using a confusion matrix to measure the accuracy of the system in classifying the average accuracy using a linear kernel. The results of the classification obtained by Anies Baswedan get an accuracy of 78.28% and a precision of 81.298%. Ganjar Pranowo received an accuracy of 82.494% and a precision of 85.642%. Prabowo Subianto received an accuracy of 83.904% and a precision of 86.22%.