<![CDATA[Social media has become the main platform for the public to express views and opinions on various events, including the appointment of Kaesang Pangarep as General Chair of the Indonesian Solidarity Party (PSI). This research aims to classify public sentiment towards the appointment using the Support Vector Machine (SVM) method with the Term Frequency-Inverse Document Frequency (TF-IDF) approach. Data was collected from Twitter using the keyword "Kaesang PSI" as well as external data on topics related to Covid-19. In the kaeasang data, 300 data were taken with each label (positive, neutral, negative) to get 100 tweets and added external data of 900 data with each label (positive, neutral, negative) to get 300 tweets. After the text preprocessing process which includes case folding, stopword removal, and stemming. The model was tested using a confusion matrix to evaluate performance based on accuracy, precision, recall and F1 Score metrics. The results show that the SVM model with TF-IDF has an F1 Score of 0.53, accuracy of 0.62, precision of 0.52, and recall of 0.57. Adding external data related to Covid-19 to the TF-IDF feature has been proven to significantly improve model performance. In conclusion, the SVM method with TF-IDF is effective in analyzing sentiment on social media even with small datasets.]]>
Copyrights © 2024
