<![CDATA[The massive spread of hoax news online, particularly on the topic of natural disasters, has become a significant problem causing public unrest and anxiety. This research aims to design and build a system capable of automatically detecting hoax news to help improve digital literacy. The methodology involved collecting news data from online portals using web scraping techniques, followed by data preprocessing which included case folding, tokenization, stopwords removal, and stemming. The Term Frequency-Inverse Document Frequency (TF-IDF) method was used for feature representation. A classification model was built using the Multinomial Naïve Bayes algorithm and subsequently implemented as a functional web application using the Django framework. The model's performance evaluation demonstrated excellent results, achieving an Accuracy of 97.26%, with balanced Precision, Recall, and F1-Score values of 0.97 for both hoax and valid classes. This study concludes that the Naïve Bayes-based hoax detection system, implemented as a web application, is an effective and viable solution for assisting users in the initial verification of online news.]]>
Copyrights © 2025
