Methods and Applications of Graph Neural Networks for Fake News Detection Using AI-Inspired Algorithms

Abstract

Graph data, which often includes a richness of relational information, are used in a vast variety of instructional puzzles these days. Modelling physics systems, detecting fake news on social media, gaining an understanding of molecular fingerprints, predicting protein interfaces, and categorising illnesses all need graph input models. Reasoning on extracted structures, such as phrase dependency trees and picture scene graphs, is essential research that is necessary for other domains, such as learning from non-structural data such as texts and photos. These types of structures include phrase dependency trees and image scene graphs. Graph reasoning models are used for this kind of investigation. GNNs have the ability to express the dependence of a graph via the use of message forwarding between graph nodes. Graph convolutional networks (GCN), graph attention networks (GAT), and graph recurrent networks (GRN) have all shown improved performance in response to a range of deep learning challenges over the course of the last few years.