The application scenarios of false information detection are extensive, involving multiple fields such as news media, social networks, and e-commerce. In the news media field, false information detection can help news organizations and reporters distinguish and block fake news, improving the credibility of news reporting. In the social network field, false information detection can help social network platforms discover and remove false information in a timely manner, thereby maintaining the health and order of the network space. In the e-commerce field, false information detection can help consumers identify fake goods and false advertising and safeguard the rights and interests of consumers.

In false information detection tasks, there are many research results available for reference. Traditional content-based approaches, which analyze the credibility of individual tweets or claims separately, ignore the high correlation between tweets and events and do not consider information from human–content interaction data. In recent years, many methods based on deep learning and methods based on novel feature fusion have also emerged. For example, Ma proposed a method for rumor detection using a tree-structured recursive neural network, and the results show that the proposed method can achieve excellent early detection of rumors [2]. However, this method may have some difficulties in processing long texts and complex syntax structures. Vu proposed a new model based on graph convolutional networks and propagation embedding for rumor detection in social media and conducted sufficient experiments on real datasets to prove the effectiveness of the method [3]. However, this method is aimed at text propagation in social networks and may not be suitable for cases involving visual or other non-textual information in the propagation process. In addition, Monti proposed a new type of automatic fake news detection model based on geometric deep learning, and experiments showed that social network propagation and structure are important features for the highly accurate detection of fake news [4]. However, this method may be affected by the sparseness and incompleteness of the data. Social media data often have high dynamicity and noise and lack complete information labels, which may lead to a decline in the performance of the model. In addition, since different news data are often interrelated, they are naturally interactive. Graph neural networks that are good at processing graph structures are also applied to rumor detection tasks. For example, Lotfi proposed a model that uses graph convolutional networks to detect rumor conversations, which extracts reply trees and user graphs for each conversation, achieving better performance compared to the baseline method [5]. Bian proposed a bidirectional graph model called a bidirectional graph convolutional network (Bi-GCN) to explore both features through top-down and bottom-up rumor propagation. It uses a GCN with a top-down directed graph of rumor propagation to learn the patterns of rumor propagation, and a GCN with a rumor propagation graph in the opposite direction to capture the structure of rumor propagation, which empirically demonstrated the superiority of this method over state-of-the-art methods [6]. Qian proposed a hierarchical multimodal contextual attention network (HMCAN) that uses the Resnet and Bert models to learn the features of images and text, respectively, and designed a hierarchical coding network to capture hierarchical semantics for fake news detection [7].

Traditional methods for detecting fake information can be categorized into two types: rule-based methods and machine learning-based methods. Rule-based methods classify real and fake information using the differences between them, including features like keywords, sentence structures, and sentiment polarities in the text [8]. On the other hand, machine learning-based methods classify real and fake information by building models, such as using support vector machines, random forests, or other algorithms to train and classify data. Ma et al. first used deep learning models to detect rumors on Weibo [9]. In subsequent studies, they proposed two recursive neural network models based on top-down and bottom-up tree structures to better capture rumor structures and text features. The results showed that this model achieved high accuracy in detecting early propagating rumors [2].

Since introducing deep learning methods, the accuracy and efficiency of fake information detection have significantly improved. Among them, graph neural networks, as a powerful representation learning method, have a wide range of applications in fake information detection. Graph neural network-based fake information detection methods usually fall into two categories: node-based methods and graph-based methods.

Node-based methods mainly focus on node features and contextual information, whereas graph-based methods use the entire graph as input and utilize graph neural networks to learn graph representations. In node-based methods, the most commonly used approach is to use the social and content attributes of nodes for fake information detection, for example, by using features such as user information, text content [2,10], and time information to determine whether the node is spreading fake information. Y Liu et al. used recurrent and convolutional networks to construct a time-series classifier to capture global and local variations in user features on the propagation path, thus detecting fake news [11]. This method is the first to model the news dissemination path on social media as multi-dimensional time series and practice fake news detection through a sequence classifier. Ling Sun et al. discussed a novel joint learning model called HG-SL for the early detection of fake news. This model uses a hyper-GNN to embed the global relationships of users, and multi-head self-attention modules to simultaneously learn local contexts (local context in specific news) during propagation in order to comprehensively capture the differences between true and false news. The introduction of global node centrality and local propagation status further highlights user influence and news dissemination ability. The experiments show that HG-SL is significantly better than the SOTA models in the early detection of fake news [12]. In addition, some studies have also considered the propagation behavior of nodes as one of the node features, such as the forwarding and like counts of nodes [13,14].

In graph-based methods, the main approach is to learn the representation of the entire graph through graph neural networks and then perform fake information detection. For example, Tian Bian et al. proposed a novel bidirectional graph convolutional network (Bi-GCN) model, which uses a rumor propagation-directed graph with a top-down structure to learn the propagation patterns of rumors, and a rumor propagation graph with a reverse direction to capture the structure of rumor propagation. The influence of the original post of the rumor is enhanced in the graph structure, and the model achieved excellent results in fake information detection [6]. K Tu et al. proposed a framework for rumor representation learning and detection. This framework uses combined text and propagation structure representation learning to improve rumor detection performance. The authors proposed a joint graph concept to integrate the propagation structure of all tweets to alleviate the sparsity issue of the propagation structure in the early stage [15]. Some researchers have also combined graph neural networks with attention mechanisms to learn graph representations more accurately. For instance, Qi Huang et al. proposed a meta-path-based heterogeneous graph attention network framework. The heterogeneous graph is decomposed into tweet word and user subgraphs according to tweet words and tweet user paths, and node representations are learned using subgraph attention networks to capture the global semantic relationships of text content and fuse information involved in source tweet propagation for rumor detection [16]. Chunyuan Yuan et al. proposed a novel global-local attention network (GLAN) for rumor detection on social media. Their method combines local semantic relationships with global structural information, uses multi-head attention mechanisms to integrate the semantic information of relevant retweets into the source tweet, generates a better-integrated representation, establishes a heterogeneous graph using global structural information to capture the complex global information between different source tweets, and uses global attention for rumor detection. Experimental results show that the GLAN is significantly better than existing models in rumor detection and early detection [17]. In addition, Ma et al. used statistical features in three aspects, including rumor content language characteristics, user characteristics participating in rumor transmission, and the propagation network structure to build a feature graph. The authors integrated entity recognition, sentence reconstruction, and ordinary differential equation networks into a unified framework called ESODE, which improved the performance of rumor detection [18].