<![CDATA[Detecting fake news has become increasingly important in the digital era, where false information can spread rapidly and significantly influence public opinion. The dissemination of fake news can lead to public distrust in the media, economic losses, and even social conflict. This study aims to develop an effective fake news detection system using the Random Forest algorithm approach. The dataset used in this research was collected from the official Kominfo website and includes attributes such as title, description, author, date, category, page, news URL, and image URL. The text preprocessing process involves tokenization, stop word removal, text normalization, and feature extraction using Term Frequency and Inverse Document Frequency (TF-IDF) to generate numerical representations of the textual data. The Random Forest model was evaluated using accuracy, precision, recall, and F1-score metrics to assess its effectiveness in detecting fake news. The results show that the model performed exceptionally well, with k-fold cross-validation (k=5) yielding high average accuracy—Random Forest achieved an accuracy of 0.9890.]]>
Copyrights © 2025
