COMPARISON OF NAÏVE BAYES CLASSIFIER, SUPPORT VECTOR MACHINE, RANDOM FOREST ALGORITHMS FOR PUBLIC SENTIMENT ANALYSIS OF KIP-K PROGRAM ON TWITTER

The Kartu Indonesia Pintar Kuliah (KIP-K) program has become a hot topic of conversation on social media Twitter, with various public sentiments regarding its implementation. The program is regulated through Minister of Education and Culture Regulation (Permendikbud) No. 10/2020, which notes an increase in the number of recipients from 552,706 in 2020 to 985,577 in 2024. However, controversy has arisen due to the alleged misuse of KIP-K funds by some influencers to support lavish lifestyles. This study aims to compare the performance of Naive Bayes, Support Vector Machine, and Random Forest algorithms in classifying public sentiment towards the KIP-K program. The research dataset was obtained from Twitter with a total of 6,842 tweets collected using crawling techniques in the time span of 2023 to 2024. The dataset was then processed through the preprocessing stage to produce clean data. The three algorithms were tested to evaluate the accuracy of each model in predicting public sentiment. The test results show that the Random Forest algorithm has the best performance with 100% accuracy, followed by Support Vector Machine with 99% accuracy, and Naive Bayes with 91% accuracy after optimization (SMOTE). Based on these findings, Random Forest proved to be the most effective algorithm in classifying sentiments related to the KIP-K program. It is hoped that the results of this research can help the management of the KIP-K program to be more targeted by providing a better understanding of public perception.