<![CDATA[The rapid advancement of information technology has enabled the widespread dissemination of information through digital platforms, but it has also led to the spread of false or hoax news. This study aims to develop a text-based news classification system using the Naïve Bayes Classifier algorithm to detect and distinguish between hoax and non-hoax news. The research uses the Knowledge Discovery in Databases (KDD) method, which includes data selection, preprocessing (case folding, tokenization, stopword removal, and stemming), feature extraction with TF-IDF, and classification using Naïve Bayes. Evaluation is performed using a confusion matrix to measure accuracy, precision, recall, and F1-score. The resulting model achieved an accuracy of 83.15%. It showed a high recall of 94% for the hoax class, indicating strong performance in identifying hoax news. However, the recall for non-hoax news was only 60%, showing limitations in detecting legitimate news. Precision was relatively balanced at 83% for hoax and 82% for non-hoax. Overall, the Naïve Bayes algorithm proved to be fairly effective in building a text-based hoax news detection system and is expected to serve as an initial solution to help the public verify the truth of information automatically and efficiently.]]>
Copyrights © 2026
