Spreading phenomena on complex networks and social systems

• Ano de defesa: 2020
• Autor(a) principal: Cota, Wesley Francis Costa
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa
• Palavras-chave em Português: Teoria das redes complexas (Física); Fenômenos críticos; Sociofísica; Física Estatística e Termodinâmica
• Link de acesso: https://locus.ufv.br//handle/123456789/28494

Epidemic spreading has been one of the most prominent and widely investigated issues in the recent literature of complex networks. Despite several advances, the mathematical modeling of spreading processes remains the target of intensive investigations. In this thesis, we initially focused on the fundamental aspects of these dynamical processes on large and highly hetero- geneous networks. We investigate some modifications of the susceptible-infected-susceptible (SIS) epidemic model to determine the robustness of epidemic thresholds by using extensive computer simulations and mean-field theories. Our results suggest that the metastable states of the standard SIS dynamics are not universal on networks that follow a power-law degree distri- bution P(k) ∼ k −γ , with γ > 5/2. We also investigated the effects of quenched disorder in the SIS dynamics on modular networks, whose results point to the existence of extended regions of criticality in networks with infinite dimension and lack of hierarchy, leading to Griffiths phases, which can be a mechanism to explain the criticality of complex systems such as the brain. We also focused on more specific problems of spreading processes involving social and human mobility patterns. First, we explored the influence of political lean in information spreading over a political communication network reconstructed using data collected from Twitter. In this case, we showed that although opposite sides of a discussion form the so-called echo cham- bers, sharing similar opinions with each other, some users were able to spread the information better than others, breaking these echo chambers. This spreading capacity was strongly re- lated to a diversity of users reached. Finally, we also examined the epidemic spreading on top of metapopulations. We adapted an existing analytical framework for the SIS dynamics on populations with recurrent mobility to accommodate the heterogeneous nature of human social contacts. We derived an analytical expression that involves demography, mobility, and contact patterns, that were used in synthetic networks to understand how they can change the epidemic threshold. All these contributions take a step further to a better understanding of the relationship between the structure of networks and epidemic dynamics, which is extremely important to improve epidemic models to deal with real situations. Keywords: Complex networks. Spreading processes. Critical phenomena. Sociophysics.