Umeå University's logo

umu.sePublications
EnglishSvenskaNorsk
Change search
ExportLink to record
Permanent link

Direct link
BETA

Project

Project type/Form of grant
Grant for employment or scholarship
Title [sv]
Ett mesoskop för komplexa system: Kartläggning av flödesvägar i sociala och biologiska system
Title [en]
A mesoscope for complex systems: Mapping flow pathways in social and biological systems
Abstract [en]
The outbreak of swine flu in 2009 killed more than 200,000 people globally and is unfortunately not the last pandemic to hit us. Moreover, we are causing the largest extinction of species since the loss of the dinosaurs 65 million years ago, with severe consequences on biodiversity and ecological resilience. To understand and mitigate pandemic outbreaks, range shifts of species, and other important phenomena of complex systems, researchers from different areas of science rely on tools from network science.Founded in statistical physics, these tools model sequential movements of some entity between various components of the complex system as flows on networks to reveal system function --- such as how our travel patterns affect the spreading of flu. Today researchers rely on one-step models of flows on networks, but this conventional approach ignores the fact that the flow direction often depends on two or more steps, that is, where the flows come from --- thereby erroneously assuming, for example, that Americans and Asians visiting Europe are equally likely to fly to Asia.Moreover, conventional approaches also aggregate different types of flows over time into a single network, inevitably destroying multilayer and temporal information, such as ignoring that transit passengers arrive before they depart. However, recent evidence suggests that such higher-order information about real flow pathways is critical for capturing all-important phenomena in the dynamics and function of the system, including its organization and spreading efficacy.This evidence exposes a serious shortcoming of conventional approaches and raises a major scientific challenge: we need to move beyond one-step models on networks and capture and comprehend the higher-order effects of real flow pathways in a barrage of data in order to understand the continuously changing organization of social and biological systems.The project builds on a an important discovery I recently made. Mapping higher-order flows was completely unexplored research terrain until I showed how functional flow modules, the building blocks of complex systems, can be revealed in two-step flows. While this approach so far only works for small systems, and a second-order model is insufficient to fully capture real flow pathways, I have just shown that what researchers have considered as two different challenges, mapping multi-step flows andmultilayer networks, are in fact two aspects of the same challenge. This breakthrough discovery enables scalable mapping of flow pathways based on all regularities in the data.The overall aim of this project is to capitalize on my pioneering work on mapping flow pathways and develop novel algorithms and tools that take advantage of today's data explosion for revealing important organizational structures in complex social and biological systems. This will allow us to address pressing research questions concerning seasonal outbreaks of flu and changes in geographical organization of life on Earth. Because of the pervasiveness of network modeling in all areas of science, the both conceptually and practically important advances of this project will have broad and significant impact. The mapping tools will make use of increasingly available rich interaction data and allow us to attack specific research questions in multidisciplinary collaborations across the sciences.In the six-year project, we will focus on two key research questions with important practical implications:(i) Better understanding of variation in seasonal flu outbreaks will allow us to make better forecasts and thereby suggest more efficient interventions to mitigate outbreaks.(ii) Better methods for identifying human-driven changes in the geogra
Principal InvestigatorRosvall, Martin
Coordinating organisation
Umeå University
Funder
Period
2017-01-01 － 2022-12-31
National Category
Other Physics TopicsEcologyPublic Health, Global Health, Social Medicine and Epidemiology
Identifiers
DiVA, id: project:1487Project, id: 2016-00796_VR
v. 2.46.0
|
WCAG
|
Umeå University Library
|
Register in DiVA
|
Support
|
Search DiVA Portal