Multilingual Content Moderation for Health Misinformation on Social Media

Published Paper PDF: View PDF

Confirmation Letter: View

Rakesh Pal

Independent Researcher

Salt Lake Sector V, Kolkata, India (IN) – 700091

Abstract— Health misinformation spreads rapidly across global social media ecosystems that operate in dozens of scripts and hundreds of dialects. While platform policies and automated moderation have matured in English, performance remains uneven across non-English and code-switched content, leaving large populations underprotected. This manuscript proposes a comprehensive, multilingual content-moderation framework tailored to health misinformation. We synthesize current evidence on infodemic management and automated detection, highlight data gaps that limit cross-lingual generalization, and describe a practical pipeline that combines language identification and normalization, code-switch detection, cross-lingual semantic representations, health-claim extraction and grounding, risk-aware triage, and human-in-the-loop mechanisms. To make the discussion concrete, we outline an illustrative evaluation using public datasets (e.g., CoAID, ReCOVery, FakeHealth, FakeCovid, MM-COVID) and small hand-labels for Hindi and Spanish. A single summary table reports plausible, sample results and shows how the proposed pipeline improves macro-F1 and reduces harmful false negatives versus a multilingual baseline. Statistical procedures (paired bootstrap for F1, McNemar’s test for error rate differences, DeLong’s test for AUC) are specified to enable reproducible analysis. We close with guidance for governance, transparency, and equity auditing so that multilingual communities—especially low-resource language users—receive timely, accurate health information without undue over-moderation risk.

Keywords— Health Misinformation, Multilingual Moderation, Cross-Lingual NLP, Infodemic Management, Social Media, Code-Switching, Risk Triage, Human-in-the-Loop

References

Adebesin, F. (2023). The role of social media in health misinformation and disinformation: Challenges and opportunities. Information Services & Use, 43(4), 515–530. https://doi.org/10.3233/ISU-230170 PMC
Barve, Y., Shukla, N., Sampath, H., & others. (2023). An automated fact-checking-based approach to combating health misinformation. Journal of Medical Internet Research, 25, e39661. PMC
Cheng, M., Nazarian, S., & Dagan, I. (2021). A COVID-19 rumor dataset. Data in Brief, 36, 107032. PMC
Cui, L., & Lee, D. (2020). CoAID: COVID-19 healthcare misinformation dataset. arXiv:2006.00885. arXiv
Dai, E., Sun, Y., & Wang, S. (2020). Ginger cannot cure cancer: Battling fake health news with a comprehensive data repository (FakeHealth). arXiv:2002.00837. arXiv
Kbaier, D., Achouri, S., Sas, C., & Tricart, J. (2024). Prevalence of health misinformation on social media: A scoping review. JMIR Infodemiology, 4(1), e51605. PMC
Kim, J., Tabibian, B., Oh, A., Schölkopf, B., & Gomez-Rodriguez, M. (2021). FibVID: Fake news diffusion dataset during COVID-19. Journal of Informetrics, 15(4), 101206. ScienceDirect
Kolluri, N., et al. (2022). COVID-19 misinformation detection: Machine-learned approaches and challenges. JMIR Infodemiology, 2(2), e38756. jmir.org
Li, Y., Shu, K., & others. (2020). Toward a multilingual and multimodal data repository for COVID-19 fake news (MM-COVID). IEEE BigData Workshops. emory.edu
Li, Y., Zhang, H., & Wang, X. (2025). Enhancing health misinformation detection with deep learning. Information Processing & Management, 62(5), 103687. ScienceDirect
Nickel, B., Hersch, J., & Barratt, A. (2025). Social media posts about medical tests with potential for overdiagnosis or overuse. JAMA Network Open, 8(2), e250123. JAMA NetworkPMC
Nasser, M., et al. (2025). A systematic review of multimodal fake news detection on social media. Journal of Information Security and Applications, 80, 103823. ScienceDirect
Özçelik, O., Güngör, T., & Pamuksuz, U. (2023). Cross-lingual transfer learning for misinformation detection. Proceedings of LDK 2023, 543–552. ACL Anthology
Rodrigues, F., et al. (2024). The social media infodemic of health-related misinformation. Journal of Behavioral and Experimental Economics, 104, 102039. ScienceDirect
Schlicht, I. B., Lehmann, R., & others. (2023). Automatic detection of health misinformation: A systematic review. PLOS Digital Health, 2(5), e0000207. PMC
Shahi, G. K., & Nandini, D. (2020). FakeCovid: A multilingual, cross-domain fact-checked news dataset for COVID-19. arXiv:2006.11343. arXiv
(2022). Policy brief: COVID-19 infodemic management. World Health Organization. WHO IRIS
(2023). Global research and innovation for health emergencies: Infodemic management priorities 2023–2024. World Health Organization. World Health Organization
WHO/Europe. (2024). Advancing infodemic management in RCCE: Implementation guidance. World Health Organization Regional Office for Europe. WHO IRIS
Zhou, X., Mulay, A., Ferrara, E., & Zafarani, R. (2020). ReCOVery: A multimodal repository for COVID-19 news credibility research. CIKM ’20 Proceedings, 3205–3212.