Enriching GNNs with Text Contextual Representations for Detecting Disinformation Campaigns on Social Media
This addresses disinformation detection for social media platforms, but is incremental as it builds on existing GNN approaches with improved text features.
The paper tackled fake news detection on social media by incorporating Transformer-based textual features into Graph Neural Networks, achieving a 33.8% relative improvement in Macro F1 over models without textual features and 9.3% over static text representations.
Disinformation on social media poses both societal and technical challenges, requiring robust detection systems. While previous studies have integrated textual information into propagation networks, they have yet to fully leverage the advancements in Transformer-based language models for high-quality contextual text representations. This work addresses this gap by incorporating Transformer-based textual features into Graph Neural Networks (GNNs) for fake news detection. We demonstrate that contextual text representations enhance GNN performance, achieving 33.8% relative improvement in Macro F1 over models without textual features and 9.3% over static text representations. We further investigate the impact of different feature sources and the effects of noisy data augmentation. We expect our methodology to open avenues for further research, and we made code publicly available.