Detecting Fake News in Social Media Networks

TLDR

Fake news, defined as deliberately fabricated articles, has existed long before the internet and is now widely disseminated on social media and news outlets, often as clickbait to generate advertising revenue. The study aims to analyze fake news prevalence on social networks and develop a user‑usable tool to detect and filter misleading sites. The authors employ a lightweight feature set extracted from titles and posts to train a logistic classifier for fake news detection. The approach achieves 99.4% accuracy in classifying fake posts.

Abstract

Fake news and hoaxes have been there since before the advent of the Internet. The widely accepted definition of Internet fake news is: fictitious articles deliberately fabricated to deceive readers”. Social media and news outlets publish fake news to increase readership or as part of psychological warfare. Ingeneral, the goal is profiting through clickbaits. Clickbaits lure users and entice curiosity with flashy headlines or designs to click links to increase advertisements revenues. This exposition analyzes the prevalence of fake news in light of the advances in communication made possible by the emergence of social networking sites. The purpose of the work is to come up with a solution that can be utilized by users to detect and filter out sites containing false and misleading information. We use simple and carefully selected features of the title and post to accurately identify fake posts. The experimental results show a 99.4% accuracy using logistic classifier.