Sentiment Analysis Of Indonesian Capital Movement Using The Naive Bayes Algorithm

Abstract

Using the hashtags #IKN and #IbuKotaPindah, this study aims to ascertain how Indonesians on Twitter feel about the relocation of the capital. The data is then processed by text processing after being saved in a CSV file. By using the Nave Bayes Classifier method, the text classification process is split into positive and negative sentiment classes. The accuracy of the algorithm is determined by the test results in analyzing the performance of the algorithm using the Confusion Matrix reference and expressing the predictions and actual conditions of the data generated by the algorithm. The f1-score is the average of precision and recall, which are expressed as the positive class at 79 percent and the negative class at 96 percent. Precision is the ratio of the correct amount predicted on the positive class label at 82 percent, negative 95 percent, and recall is the ratio of the correct predicted amount to the overall correct data. Rapid Miner 9.10 is used. Display tab visualizations of the histogram, wordcloud, test data table, and original data table.