Enriching GNNs with Text Contextual Representations for Detecting Disinformation Campaigns on Social Media

Bruno Croso Cunha da Silva, Thomas Palmeira Ferraz, Roseli De Deus Lopes

Keywords: Graph Neural Networks, Applications, Disinformation Campaign, Language Modeling, Noisy Augmentation, Embeddings, Transformer

TL;DR: We investigate how incorporating contextual text features into Graph Neural Networks (GNNs) improves fake news detection on social media, showing a 9.3% boost in Macro F1 over static embeddings and 33.8% over GNNs without text.

Abstract:

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.

Submission Type: Extended abstract (max 4 main pages).

Software: https://github.com/BrunoCroso/ContextualGNNs-FakeNews

Submission Number: 136