La mod\u00e9lisation vous passionne\u202f? Le Pr\u00e9crisa vous invite \u00e0 un webinaire anim\u00e9 par Jacques B\u00e9lair, professeur titulaire en math\u00e9matiques et statistique \u00e0 l\u2019Universit\u00e9 de Montr\u00e9al et coresponsable du groupe de travail sur la mod\u00e9lisation et les donn\u00e9es complexes du Pr\u00e9crisa. D\u00e9couvrez, avec Zitao He, comment les comportements humains peuvent \u00eatre int\u00e9gr\u00e9s dans les mod\u00e8les \u00e9pid\u00e9miologiques pour mieux comprendre et pr\u00e9voir les dynamiques de sant\u00e9 publique.<\/p>\n
R\u00e9sum\u00e9 de la pr\u00e9sentation (en anglais)<\/strong> Pr\u00e9sentateur<\/strong> Format : En ligne (Zoom)
\nUnderstanding how human behaviors impact infectious disease dynamics is critical for public health, particularly in socially and behaviorally diverse populations. The talk presents two complimentary modeling studies that examine how heterogeneity and homophily (the tendency to interact with similar others) shape vaccination decisions and epidemic outcomes. The first study considers a population divided into social media users and non-users within a coupled behavior-disease framework. We show that stronger homophily across these groups slows the spread of pro-vaccine behavior and accelerates the system\u2019s approach to a tipping point leading to low vaccine uptake. The model also highlights the role of early online discussion as a potential early warning signal for future disease outbreaks. The second study designs a more generalized multi-group vaccination game to explore how social groups with different baseline vaccine support levels influence one another. We demonstrate that strong homophily can produce persistent opinion polarization, even in the presence of severe disease, and that mixed-strategy equilibria can emerge across all groups. Together, these studies illustrate how incorporating social and behavioral heterogeneity into epidemic models can improve our understanding of real-world vaccination dynamics and inform more targeted public health strategies. The talk will emphasize conceptual insights, modeling approaches, and implications for interdisciplinary research at the interface of mathematical modeling and social science.<\/p>\n
\nZitao He est r\u00e9cemment titulaire d\u2019un doctorat en math\u00e9matiques appliqu\u00e9es de l\u2019Universit\u00e9 de Waterloo, o\u00f9 il a men\u00e9 des travaux de recherche sur les syst\u00e8mes dynamiques guid\u00e9s par les donn\u00e9es en \u00e9pid\u00e9miologie comportementale. Il poss\u00e8de une solide exp\u00e9rience en mod\u00e9lisation math\u00e9matique, apprentissage automatique, science des donn\u00e9es et \u00e9pid\u00e9miologie. Il a d\u00e9montr\u00e9 sa capacit\u00e9 \u00e0 combiner mod\u00e8les math\u00e9matiques et m\u00e9thodes de deep learning pour analyser des ensembles de donn\u00e9es complexes, notamment issus des m\u00e9dias sociaux et de la surveillance \u00e9pid\u00e9miologique, ainsi qu\u2019\u00e0 d\u00e9velopper des syst\u00e8mes d\u2019alerte pr\u00e9coce pour la d\u00e9tection de flamb\u00e9es de maladies. M. He est actuellement gestionnaire \u00e0 la Banque Scotia, au sein de la division de s\u00e9curit\u00e9.<\/p>\n
\nLangue : Anglais
\nLien d\u2019inscription : Inscription au webinaire – Zoom<\/a><\/p>\n