Human Behaviour & Social Impact
Epidemics are not just biological events. They are deeply social ones. At AIMMLab, we study how human behaviour, communication platforms, and public perception shape the trajectory of disease outbreaks. Using mathematical modelling, AI, and large-scale social media analysis, we generate insights that help policymakers, public health agencies, and communities navigate the dual threats of disease and misinformation.
Our published work spans Twitter-based sentiment analysis, off-label drug use modelling (including ivermectin discourse in South Africa and Nigeria during COVID-19), epidemic–economic models of human responses to public health policy, and the stigmatisation of marginalised communities during the mpox outbreak. These studies demonstrate that misinformation and behavioural dynamics are not peripheral to epidemic modelling, they are central to it.
Why this matters: Research shows that social media misinformation can increase vaccine hesitancy within days of being posted with measurable effects on uptake rates even after controlling for political and demographic factors. During COVID-19, an estimated 40% of online health-related messages were generated by automated bots. These dynamics are not noise they are signals that AIMMLab's models are built to detect and quantify.
Key Focus Areas
Behavioural Dynamics Modelling
Mathematical models of how populations adopt preventative behaviours, develop vaccine hesitancy, and comply or not with public health guidelines during epidemics and pandemics.
Misinformation & Infodemic Detection
AI-powered monitoring of social media platforms to detect misinformation outbreaks, track their spread, and quantify their impact on public health behaviour and policy compliance.
Sentiment & Public Opinion Analysis
Real-time natural language processing and transformer-based models applied to social media data to track emotional trends, risk perception, and community sentiment during health crises.
Stigma & Self-Medication Behaviour
Studying how stigmatisation of marginalised groups during mpox, or of self-medicating communities during COVID-19 shapes help-seeking behaviour, healthcare access, and public trust across diverse cultural contexts.
Epidemic–Economic Modelling
Integrating economic and behavioural variables into epidemic models to assess how policy interventions lockdowns, travel restrictions, vaccination mandates interact with human responses and disease progression.
Media Influence & Information Networks
Analysing how media framing, computational advertising, and online influence networks shape community responses during health emergencies and how these can be leveraged for effective public health communication.
Figure: AIMMLab's social-epidemic analysis pipeline — from real-time social signals to evidence-based public health action.
Outcomes
- Validated mathematical and AI models capturing the co-evolution of disease spread and human behavioural response.
- Early detection tools for misinformation outbreaks and shifts in public sentiment — deployable in real time during health crises.
- Socially-informed epidemic models that account for stigma, self-medication, hesitancy, and cultural context.
- Evidence-based recommendations for policymakers and communicators on combatting disinformation and designing equitable, culturally responsive public health campaigns.
📚 Selected Refereed Journal Publications
Sharma, Y., Laison, E. K., Philippsen, T., Ma, J., Kong, J., Ghaemi, S., ... & Nasri, B. (2024). Models and data used to predict the abundance and distribution of Ixodes scapularis (blacklegged tick) in North America: a scoping review. The Lancet Regional Health–Americas, 32.
Yuh, M. N., Ndum Okwen, G. A., Miong, R. H. P., Bragazzi, N. L., Kong JD., Movahedi Nia, Z., ... Patrick Mbah, O. (2024). Using an innovative family-centered evidence toolkit to improve the livelihood of people with disabilities in Bamenda (Cameroon): a mixed-method study. Frontiers in Public Health, 11, 1190722.
Nunes, M. C., Thommes, E., Fröhlich, H., Flahault, A., Arino, J., Baguelin, M., Kong JD ... Coudeville, L. (2024). Redefining pandemic preparedness: Multidisciplinary insights from the CERP modelling workshop in infectious diseases, workshop report. Infectious Disease Modelling.
Kaur, M., Cargill, T., Hui, K., Vu, M., Bragazzi, N. L., Kong JD (2024). A Novel Approach for the Early Detection of Medical Resource Demand Surges During Health Care Emergencies: Infodemiology Study of Tweets. JMIR Formative Research, 8, e46087.
Bain, L. E., Yankam, B. M., Kong, J. D., Nkfusai, N. C., Badru, O. A., Ebuenyi, I. D., ... & Adeagbo, O. (2023). Global Health Mentorship: Challenges and Opportunities for Equitable Partnership. BMJ Global Health, 8(11), e013751.
Kaur, M., Bragazzi, N. L., Heffernan, J., Tsasis, P., Kong JD (2023). COVID-19 in Ontario Long-term Care Facilities Project, a manually curated and validated database. Frontiers in Public Health, 11, 1133419.
Nia ZM, Ahmadi A, Mellado B, Wu J, Orbinski J, Asgary A, Kong JD. Twitter-based gender recognition using transformers. Math Biosci Eng. 2023 Aug 3;20(9):15962-15981. doi: 10.3934/mbe.2023711.
Movahedi Nia, Z., Bragazzi, N., Asgary, A., Orbinski, J., Wu, J., Kong JD (2023). Mpox Panic, Infodemic, and Stigmatization of the 2SLGBTQIA+ Community: Geospatial Analysis, Topic Modeling, and Sentiment Analysis. Journal of Medical Internet Research, 25, e45108.
Fevrier, K., Effoduh, J. O., Kong JD, Bragazzi, N. L. (2023). Artificial Intelligence, Law, and Vulnerabilities. In AI and Society. : 179-196.
Ji J, Wang H, Wang L, Ramazi P, Kong JD, Watmough J. Climate-dependent effectiveness of nonpharmaceutical interventions on COVID-19 mitigation. Mathematical Biosciences. 2023 Dec 1;366:109087.