FAKE NEWS DETECTION USING NOVEL GAN BASED TECHNIQUE

Abstract

The proliferation of fake news across digital platforms presents a critical threat to information credibility and the integrity of public discourse. This paper proposes a novel fake news detection framework based on Generative Adversarial Networks (GANs). The framework leverages GANs to synthesize news articles while training a discriminator to effectively distinguish between authentic and fabricated content. Through adversarial learning, the discriminator progressively enhances its classification capability by identifying subtle linguistic and contextual cues that separate real news from misinformation. Experimental evaluation demonstrates the robustness of the proposed model in improving detection accuracy, highlighting its potential for deployment in automated verification systems and digital media integrity solutions.