The complexity of heterogeneity in real-world networks

Many real-world phenomena are characterized by complexstructures. Modeling and detecting these architectures is ofparamount importance to understand the dynamics of theconsidered systems and to consciously intervene on them.The COVID-19 pandemic is, in this sense, a telling example.One common feature of these complex structures is theheterogeneity of the node connectivity of the underlying network.This heterogeneity is one of the main culprits of theongoing pandemic.In this thesis, we introduce a method capable of uncoveringcomplex networks’ heterogeneous structures when finite-sizeeffects hide the latter. Larger heterogeneity in the networkstructure leads to a smaller epidemic threshold. It is not a coincidencethat policymakers worldwide are trying to reducethis heterogeneity (employing lockdowns and other less restrictivemeasurements) to stop the spread of the virus. Weshow that a macro-quarantine followed by a micro-quarantinecan help in this direction. For this scope, we introduce an algorithmthat is able to track super-spreaders. Notably, thesame algorithm can be used to define an optimized strategyfor vaccinations. Similarly to a virus, information continuouslyspreads on social networks. By combining MachineLearning and network theory techniques, we develop an algorithmable to discover what “social users” think about aparticular topic—a sort of social listener, an AI alternative totraditional polls.