<![CDATA[Conversations on social media Twitter related to tolerance among religious communities in Indonesia are fascinating. However, it is a sensitive issue. In reality, there is often a war of comments about the implementation of tolerance between religious people in carrying out their own beliefs. The community is not careful in issuing opinions that can result in social insecurity, insecurity, and national instability. This condition will significantly affect the state of the country's economy. In some cases, political problems can be a trigger for intolerance between religious communities. The purpose of this study is to compare the performance of classification accuracy on positive or negative sentiments from conversations that intersect with the problem of tolerance among religious communities during the past year. In this study, we compared the performance of the accuracy of the modeling of sentiment analysis classification on public conversations on social media Twitter related to tolerance between religious communities in Indonesia. Because the text that will be carried out modeling comes from the Indonesian language, to facilitate labeling, translation is carried out into English, then a performance comparison of the sentiment analysis classification modeling with SVM algorithm, Naïve Bayes, Decision Tree, and k-NN. Based on the experiments, it was concluded that the SVM algorithm has the highest performance for the classification of sentiment analysis categories up to 65.03% compared to the Naïve Bayes algorithm, which reached 59.92%, Decision Tree, which reached 63.52% and k-NN which reached 57 66%.]]>
Copyrights © 2020
