Online News Hoax Detection Using Machine Learning Classification Algorithms

Abstract

The rapid growth of digital media usage has significantly increased the spread of hoax news. Such information can lead to misinformation, social anxiety, and public misunderstanding. This study proposes an automatic detection approach for Indonesian-language hoax news using machine learning-based classification algorithms. A dataset consisting of 3,000 Indonesian news articles collected from social media platforms and online news portals was employed and validated using a fact-checking website (TurnBackHoax.id). The proposed method involves text preprocessing, feature extraction using Term Frequency–Inverse Document Frequency (TF-IDF), and classification using Naive Bayes and Support Vector Machine (SVM) algorithms. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Experimental results indicate that the SVM algorithm achieves better performance than Naive Bayes in detecting hoax news. The findings demonstrate that machine learning-based classification can provide an effective solution for automatic hoax detection and can be further developed for practical implementation.