Sentiment Analysis of Independent Campus Policy on Twitter Using Support Vector Machine and Naïve Bayes Classifier

Abstract

Merdeka Belajar Kampus Merdeka (MBKM) is a program that was inaugurated by the Ministry of Education and Culture in 2020 which emphasizes the independence and independence of learning. One of the social media that gives many opinions on this policy is Twitter. The sentiments written by the public about the independent campus policy can be analyzed and categorized as positive or negative sentiments as material for review. In this research, sentiment analysis on the independent campus policy was carried out with the support vector machine algorithm and naïve Bayes classifier. Sentiment analysis begins by crawling data on Twitter in the period from November 20, 2021 to December 19, 2021, with a total of 5980 data. Then preprocessing the data is carried out to normalize and clean the data before data classification is carried out. Data that has gone through preprocessing is then labeled using Vader. Furthermore, word vectorization was carried out with TF-IDF and data classification to test the accuracy of sentiment analysis with the support vector machine algorithm and naïve Bayes classifier. The test results for 20 times show that the highest level of accuracy is obtained by the support vector machine algorithm with an accuracy of 73.12%.