THE EFFECT OF MACHINE LEARNING ALGORITHMS ON HOAX DETECTION ON SOCIAL MEDIA: IMPLICATIONS FOR NATIONAL INFORMATION SECURITY

Abstract

The spread of hoaxes on social media has become a serious threat to national information security, considering the large number of people who depend on social media as a source of information. This misinformation not only has an impact on public perception but also disrupts social and political stability. This study aims to test the effectiveness of machine learning algorithms, especially Natural Language Processing (NLP), Neural Networks, and Decision Tree, in detecting hoaxes on social media and analyzing their implications for national information security. The method used is a quantitative approach with experimental and comparative analysis of the three algorithms. The data was collected through web scraping from social media platforms and analyzed using the Confusion Matrix to assess accuracy, precision, recall, and F1-score. The results showed that NLP had the highest accuracy, reaching 92.7%, followed by Neural Networks with 90.1% and Decision Tree at 86.3%. In addition, the increase in hoax detection is directly proportional to the decrease in security incidents related to disinformation, indicating the important role of machine learning algorithms in maintaining national information stability. These findings support the implementation of hoax detection algorithms as part of a more comprehensive information security policy.