Methods for detecting false information based on graph neural networks in social media
社交媒体中基于图神经网络的虚假信息检测方法

摘要

随着社交媒体平台的快速发展，虚假信息的传播速度和范围显著增加，对社会产生了深远的影响。本研究聚焦于利用图神经网络（GNNs）技术来识别社交媒体中的虚假信息，提出了一种基于图结构数据建模的方法，通过捕捉用户互动模式及内容特征之间的复杂关系来提高检测准确性。该方法在处理大规模数据时展现出优越的性能，并能有效应对传统方法难以解决的问题。结合深度学习与图分析技术为打击虚假信息提供了新的视角和解决方案。

Abstract

With the rapid development of social media platforms, the spreading speed and scope of false information have significantly increased, exerting a profound impact on society. This study focuses on utilizing Graph Neural Networks (GNNs) technology to identify false information in social media. A method based on the modeling of graph-structured data is proposed. By capturing the complex relationships between user interaction patterns and content features, the detection accuracy is improved. This method demonstrates superior performance when dealing with large-scale data and can effectively address issues that are difficult to solve with traditional methods. The combination of deep learning and graph analysis techniques provides a new perspective and solution for combating false information.