Fake News Quick Detection on Dynamic Heterogeneous Information Networks

TL;DR

This paper introduces DHGNN, a dynamic heterogeneous graph neural network that rapidly detects fake news by modeling evolving news and author relationships, leveraging semantic representations and efficient propagation techniques.

Contribution

The paper presents a novel DHGNN model that incorporates dynamic graph structures and semantic content analysis for quick fake news detection, improving efficiency and effectiveness.

Findings

DHGNN outperforms existing GNN models in accuracy.

DHGNN reduces training time through dynamic propagation methods.

Experiments on real datasets validate the model's superior performance.

Abstract

The spread of fake news has caused great harm to society in recent years. So the quick detection of fake news has become an important task. Some current detection methods often model news articles and other related components as a static heterogeneous information network (HIN) and use expensive message-passing algorithms. However, in the real-world, quickly identifying fake news is of great significance and the network may vary over time in terms of dynamic nodes and edges. Therefore, in this paper, we propose a novel Dynamic Heterogeneous Graph Neural Network (DHGNN) for fake news quick detection. More specifically, we first implement BERT and fine-tuned BERT to get a semantic representation of the news article contents and author profiles and convert it into graph data. Then, we construct the heterogeneous news-author graph to reflect contextual information and relationships. Additionally, we adapt ideas from personalized PageRank propagation and dynamic propagation to heterogeneous networks in order to reduce the time complexity of back-propagating through many nodes during training. Experiments on three real-world fake news datasets show that DHGNN can outperform other GNN-based models in terms of both effectiveness and efficiency.