Deep Learning-Powered Approaches for News Authenticity Verification and Fake News Detection: A Comprehensive Survey

Abstract

In today’s digital world, increase of Fake News rapidly spreads substantial threats to both the public and individuals, emphasizing the crucial necessity for better-quality News Authenticity confirmation. This study gives a summary of some of the detection of bogus news using a Deep Learning (DL) technique and classification-based authenticity prediction techniques that to a large extent applied in several ways to news detection applications. Fake?True News Detection is currently a hard subject that is attracting investigation due to its detrimental effects on society. Deep Learning is employed in the crucial and often-used field of Fake News classification. Due to its excellent classification accuracy, the DL based approach has been widely used in the classification of news. Neural Network based – Convolutional Neural Networks (CNN), Recurrent Neural Network (RNN), Long Short Term Memory (LSTM), Generative Adversarial Networks (GAN), Hierarchical Attention Network (HAN), and Graph-based CNN (GCN) are some of the Deep Learning approaches that have been taken into consideration in our study and are used for developing a variety of News Detection methods. The study also includes comparative studies on a few news detection and classification methods that have been used to various Fake News prediction issues.