Fake News Detection using Adversarial Networks

Abstract

The public's perception of events, news, and other crucial content is distorted by the prevalence of incorrect information on social media platforms. Intentionally false news can affect society behaviors including increased worry, loneliness, and feelings of inadequacy. It can also result in unpleasant internet experiences. Attacks by the opposition try to spread false information through internet information networks, for societal, political, economic, and cultural reasons. A recent study suggests a method for testing the long short-term memory (LSTM) deep learning model's performance against hostile instances produced by a transformer model. The goal of the research is to investigate ma- chine leaming algorithmic elements that can encourage the spread of false in- formation. The study also analyses the effectiveness of lengthy short- term re- current neural network techniques and generative adversarial networks in identifying erroneous information. It may be possible to create more durable intelli- gent systems by having a better grasp of the adversarial attack mechanisms in social media platforms. Systems resistant to potential vulnerabilities.