| Abstrakt: |
Nowadays, social media has become the key source of information for anyone seeking about current events across the world. This information may be fake or real news. On social media platforms, fake news negatively impacts politics, the economy, and health, and affects the stability of society. The research on fake news detection has received widespread attention in the field of computer science. There are many effective methods of fake news detection technology including natural language processing (NLP) and machine learning techniques, primarily focusing on content analysis and user behavior. While these methods have shown promise, they often fall short in capturing the complex relational and propagation patterns inherent in social networks. Fake news exhibits distinct features such as misleading headlines, and fabricated content, making its detection challenging. To address these issues, Graph Neural Networks (GNNs) have been introduced as a superior solution. GNNs are particularly effective in processing graph-structured data, allowing them to model the intricate connections and dissemination patterns of news in social networks more accurately. This study provides an overview A variety of false information and their characteristics and discusses various techniques and features used in fake news detection. As well as advanced GNN-based techniques and datasets used to implement practical fake news detection systems from multiple perspectives and future research directions. In addition, tables and summary figures help researchers understand the full picture of fake news detection. Finally, the object of this review is to help other researchers improve fake news detection models using GNNs. [ABSTRACT FROM AUTHOR] |
