Hoax Detection Through the Method of Convolutional Neural

Abstract

The term "hoax" or "fake news" refers to the deliberate propagation of false material on social media in order to confuse and mislead readers in order to achieve an economic or political purpose. Furthermore, the increasing diversity and number of participants in the sphere of news authoring and transmission have resulted in the creation of news pieces that must be acknowledged regardless of whether they are trustworthy or not. Furthermore, hoaxes might undermine Indonesian society's social and political features. In 2016, Central Connecticut University published The World's Most Literate Nations, in which Indonesia ranked 60th out of 61 countries, indicating that Indonesian media literacy still needs to improve in terms of critically evaluating information and distinguishing between fake news and valid news. Based on this description, the research will develop the Synonym-Based Data Augmentation for Hoax Detection method and Easy Data Augmentation (EDA) approach. This study yielded an accuracy of 8.81, showing that it is effective at spotting fake news.