Sentiment Analysis of Public Responses on Indonesia Government Using Naïve Bayes and Support Vector Machine

Abstract

Many people are interested in knowing how the public views President Joko Widodo's administration. Text Mining analysis can be one way to collect and analyze text data about Joko Widodo's administration and extract relevant information from the data. Data was obtained by collecting tweet data about Joko Widodo's government in 2022 on Twitter using Netlyitic. Then the Text Mining analysis of Joko Widodo's government was carried out using the Navie Bayes (NVB) classification and Support Vector Machine (SVM). This classification can be used to predict sentiment or public views of the government based on the tweets collected. Based on a case study of the classification results of President Joko Widodo using Naive Bayesian classification, we obtained a precision value of 79%, a recall value of 91% and a precision value of 82%. And by using SVM, we get 85% precision, 95% recall, and 83% precision. Due to the high accuracy, recall, and precision, it can be said that SVM classification is more accurate than NVB.