<![CDATA[The development of information technology and social media has made the distribution of information easier, but it has also increased the prevalence of fake news or hoaxes. This research aims to classify hoax and non-hoax news on social media using the Naïve Bayes algorithm with the assistance of the RapidMiner application. The data used is secondary data obtained from the Kaggle website and processed thru text preprocessing stages including tokenization, stopword removal, stemming, and TF-IDF weighting. The classification process was carried out using the Cross Validation method to measure model performance. The research results show that the Naïve Bayes algorithm has an accuracy of 90.20%, and precision values of 92.25% for the hoax class and 88.33% for the non-hoax class, with recall values of 87.78% and 92.62% respectively. These values indicate that the built classification model can easily identify hoax news. Thus, the Naïve Bayes algorithm has proven to be effective and efficient for use as a method for detecting fake news on social media. Keywords: Naïve Bayes, RapidMiner, Classification, Hoax News, Text Mining]]>
Copyrights © 2025
