The role of AI in combating misinformation: leveraging text mining and social networking analysis

Clinical Snapshot

Key Findings

• P Value: Not reported in abstract

• Effect Size: ROC-AUC = 0.9979 (near-perfect discrimination); Cohen's kappa = 0.9887 (near-perfect agreement between model predictions and ground-truth labels)

• Primary Outcome: Misinformation classification performance of the hybrid RoBERTa-GRU model on a curated Monkeypox Twitter dataset (5,787 posts)

• Nnt Or Sensitivity: Sensitivity and specificity not separately reported in abstract; ROC-AUC of 0.9979 implies high sensitivity and specificity but individual values and their clinical trade-offs are not disclosed

• Confidence Interval: Not reported — a significant methodological omission

Clinical Application

Real-world deployment feasibility is limited by: (1) the model's training on a small, English-only, single-platform dataset; (2) unresolved scalability concerns acknowledged by the authors; (3) absence of a real-time inference pipeline validation; (4) computational resource requirements of transformer-based architectures in resource-limited settings. Substantial further development and prospective validation would be required before operational deployment. In the Australian context, the Australian Government Department of Health and Aged Care, the Therapeutic Goods Administration (TGA), and the Australian Health Protection Principal Committee (AHPPC) have all identified health misinformation as a public health priority, particularly following COVID-19 and the 2022 Mpox outbreak. The RACGP has published guidance on addressing vaccine hesitancy and misinformation in primary care. While an automated misinformation detection tool could theoretically support Australian public health communication monitoring, this model has not been validated on Australian social media data, does not account for Australian-specific health discourse, and is not aligned with any current TGA, AHPPC, or eSafety Commissioner framework for online health content moderation. PBS and TGA considerations are not directly relevant to this computational study. Adoption in Australian public health practice would require independent validation, ethical review, and alignment with the Online Safety Act 2021. Potentially applicable to public health surveillance teams, health communication units, and social media platform moderators seeking automated tools to flag health misinformation during infectious disease outbreaks. Not directly applicable to individual patient care.

Abstract

Misinformation on social media can be a severe threat to social trust, safety, and health of the population, especially in times of an epidemic like the Monkeypox outbreak. This study specifically focuses on a hybrid RoBERTa-GRU architecture designed to capture both contextual semantics and temporal dependencies in social media discourse. This research presents how the combination of text mining and social network analysis enables Artificial Intelligence (AI) to support misinformation detection. The proposal of a hybrid architecture that integrates RoBERTa and GRU-based embeddings in a contextual fashion and GRU-based modelling of sequential patterns helps identify and substantiate misinformation in social media posts. Based on a curated X (formerly Twitter) dataset consisting of Monkeypox posts (5787 posts), the model provided state-of-the-art results, with ROC-AUC 0.9979 and Cohen's kappa 0.9887; standalone baselines were also surpassed. Results reveal that the proposed transformer-RNN hybrid effectively captures both semantic depth and temporal relationships in misinformation detection tasks. In addition to performance, the paper addresses the limitations of dataset bias, multilingual constraint issues, and scalability as related to cross-linguistic applicability, multimodal study, and performance in real-time and resource-limited systems. The study adds value in this emerging body of knowledge on AI-driven social media analytics by offering practical guidance for mitigating health-related misinformation online.

References

1. 1.Rohan, Arafat, Islam, M. A., Yoshi, A. M., Islam, M. M., Kabir, M. F., & Ahmed, K. R. (2023). The role of AI in combating misinformation: leveraging text mining and social networking analysis. Procedia Computer Science. https://doi.org/10.1016/j.procs.2023.08.088