Sentiment Analysis of Twitter Cases of Riots at Kanjuruhan Stadium Using the Naive Bayes Method

Abstract

Sentiment analysis is a process carried out to analyze opinions, sentiments, judgments, and emotions from the riot case at the kanjuruhan stadium. The purpose of this research is to find out public opinion about the tragedy that is currently happening at the Kanjuruhan Stadium. the data was obtained from social media Twitter using the Twitter API, then after that, an analysis was carried out. data from the results of the analysis will be classified using the Naive Bayes method. The classification process is divided into 7 (seven) stages, namely Crawling, Cleansing, pre-processing, labeling, classification, data training, and data testing. In the labeling process, data is classified into 2 (two) classes, namely the positive class and the negative class. The data obtained before the preprocessing process was 1963 tweets, after the preprocessing the data obtained was 1001 tweets. The data will be trained and tested using the naive Bayes classification method. classification results obtained precision values of 82% for negative data and 65% for positive data, recall values obtained 74% for negative data and 75% for positive data, F1-score values obtained 78% for negative data and 70% for positive data, while accuracy value obtained 74%.