Sentiment Analysis of YouTube Comments on the East Java Grant Fund Case Using the Support Vector Machine (SVM) Algorithm

Abstract

The phenomenon of corruption in Indonesia, particularly the alleged misuse of grant funds in East Java Province, has triggered various public responses, many of which have been expressed through comments on YouTube. This study aims to categorize and analyze public sentiment on this issue by applying the Support Vector Machine (SVM) algorithm. The data used consists of 1,039 relevant comments collected from several YouTube videos related to the case. The comments were processed through text pre-processing stages, which included case folding, tokenizing, text cleaning, removal of stopwords, stemming, and transformation using the Term Frequency-Inverse Document Frequency (TF-IDF) method. Next, the data was manually labeled into three sentiment categories: positive, negative, and neutral. The data was then divided into training and test data with a ratio of 80:20, and used to train and test the classification model. Evaluation was carried out using accuracy, precision, recall, and f1-score metrics. The results of the study show that the SVM algorithm performs well in classifying public sentiment, so it has the potential to be used as a tool to automatically identify public perceptions of developing social and political issues.