Biology Reference
In-Depth Information
Chapter 27
Social Networks, Contagion Processes
and the Spreading of Infectious Diseases
Bruno Gon ¸ alves
1
, Nicola Perra
1
and Alessandro Vespignani
1
,
2
1
College of Computer and Information Sciences and Bouve´ College of Health Sciences, Northeastern University, Boston, MA 02115, USA,
2
Institute for
Scientific Interchange (ISI) Foundation, Via Alassio 11/c, 10126 Torino, Turin, Italy
Chapter Outline
Introduction
515
Complex Networks and the Large-Scale Spreading of Infectious
Diseases
Network Thinking
516
521
Contagion Phenomena in Complex Social Networks
518
Conclusions and Future Challenges
524
References
525
INTRODUCTION
The characterization and understanding of contagion
phenomena is crucially dependent upon the conceptual
framework adopted to describe groups of individuals
(social agents) or entire populations in spatially extended
systems, and of the interaction and behavior of individuals
at various levels, from the global scale of mobility and
transportation flows to the local scale of individual
activities and contacts. In this context, a mathematical and
statistical modeling framework has evolved from simple
compartmental models into structured approaches in
which the heterogeneities and details of the population
and system under study are becoming increasingly
important (
Figure 27.1
). In the case of spatially extended
systems, modeling approaches have been extended into
schemes that explicitly include spatial structures and
which consist of multiple subpopulations coupled by
traveling fluxes, while the epidemic within the subpopu-
lation is described according to approximations depending
on the specific case studied
[1
a complete characterization (household, workplace, etc.)
of each individual
[17,18]
.
The above modeling approaches are 'data hungry', as
they depend on detailed information about the activity of
individuals, their interactions and movement, as well as the
spatial structure of the environment, transportation infra-
structures, traffic networks, and travel times. Although for
a long time these kinds of data were simply not available,
recent years have witnessed a tremendous progress in data
gathering thanks to the development of new informatics
tools and the increase in computational power. A contin-
uous flow of data has finally become available for scientific
analysis and study. The availability of data has allowed
highlighting of complex properties and heterogeneities,
which cannot be neglected in the epidemic description.
Although these characteristics have long been acknowl-
edged as a relevant factor in determining the properties of
dynamical processes, many real-world networks exhibit
levels of heterogeneity that were not anticipated until a few
years ago, and represent a theoretical and conceptual
challenge in our understanding of the unfolding of conta-
gion processes.
Although data availability is highlighting the limits of
our conceptual and modeling frameworks, it is also
allowing the validation of results across different modeling
approaches, mathematical techniques and approximation
schemes. Furthermore, it has been possible to push forward
the development of data-driven computational approaches
9]
.Thispatchormeta-
population modeling framework has then grown into
a multi-scale framework in which the various possible
granularities of the system (country, inter-city, intra-city)
are considered through different approximations and
coupled through interaction networks describing the flows
of people and/or animals
[9
e
16]
. At the most detailed
level, the introduction of agent-based models (ABM) has
enabled us to stretch the usual modeling perspective even
more to achieve a full description of the society by
e
