|
Keywords
|
Rumor Detection, Deep Learning, Graph Neural Networks, Social Bot Detection, Machine Learning Paradigms
|
|
Abstract
|
This study aimed to review the impact of deep learning (DL) techniques on rumor detection in social media
platforms, focusing on the distinctive features and user interactions on Twitter and Sina Weibo. We have
endeavored to compare the outcomes obtained from Recurrent Neural Networks (RNN), Convolutional Neural
Networks (CNN), and Graph Neural Networks (GNN). Beyond a cursory review of existing methods, we briefly
investigate the structure of two approaches, Graph Robot Aware (SBAG) and Graph Convolutional Rumor
Detection System (GCRES), both of which employ the Graph Neural Networks (GNN) method. These two
approaches are significant because, in addition to examining the content of rumors, they pay attention to the
pattern of their spread through Graph Neural Networks (GNN) for rumor detection. These advancements
underscore the potential of DL and GNN in addressing the challenge of rumor detection in social media and
emphasize the importance of continuing innovation in this rapidly evolving field.
|