Leveraging Artificial Intelligence to Quantify the Digital Infodemic: A Longitudinal Analysis of Pathogen-Specific Misinformation in Nigeria

Abstract

Background: The digital infodemic poses a systemic threat to the eradication of viral pathogens, particularly in regions such as Nigeria, with complex digital ecosystems. While traditional content analysis is limited by scale, Artificial Intelligence (AI) offers new possibilities for real-time infodemiology. This study utilises AI to quantify the longitudinal prevalence and social engagement of misinformation regarding HIV, Hepatitis B (HBV), Polio, and COVID-19.

Methods: We deployed an AI-driven analytical framework to process 7,234 data points from Nigerian digital news and social media (2021–2026). The study utilised Natural Language Processing (NLP) for automated thematic classification and VADER sentiment analysis to calculate Average Trust Scores for clinical interventions. Regression models were used to measure the correlation between algorithmic amplification and the social transmission rate of misinformation.

Results: AI-based sentiment scoring revealed a significant trust deficit regarding chronic viral infections (HBV/HIV) that persisted longer than misinformation about acute pathogens. Automated keyword analysis identified a high volume of traditional/unverified cure mentions, achieving a social reach 40% higher than that of official public health communiqués.

Conclusion: By leveraging AI, this study establishes the first data-driven baseline for the Nigerian digital infodemic. The findings demonstrate that AI is not just a tool for analysis but a necessary component of modern public health surveillance to combat vaccine hesitancy and promote evidence-based virological treatment.