Diffusion on dynamic contact networks with indirect transmission links

Abstract

Modelling diffusion processes on dynamic contact networks is an important research area for epidemiology, marketing, cybersecurity, and ecology. However, current diffusion models cannot capture transmissions occurring for indirect interactions. For example, an airborne infected individual releases infectious particles at locations that can suspend in the air and infect susceptible individuals arriving even after the infected individual left. Thus, current diffusion models miss transmissions during indirect interactions. In this thesis, a novel diffusion model called the same place different time transmission based diffusion (SPDT) is introduced to take into account the transmissions through indirect interactions. The behaviour of SPDT diffusion is analysed on real dynamic contact networks and a significant amplification in diffusion dynamics is observed. The SPDT model also introduces some novel behaviours different from current diffusion models. In this work, a new SPDT graph model is also developed to generate synthetic traces to explore SPDT diffusion in several scenarios. The analysis shows that the emergence of new diffusion becomes common thanks to the inclusion of indirect transmissions within the SPDT model. This work finally investigates how diffusion can be controlled and develops new methods to hinder diffusion.

PhD Thesis, Macquarie University

References in corpus (14)

• Understanding individual human mobility patterns
• Finding community structure in networks using the eigenvectors of matrices
• Efficient Immunization Strategies for Computer Networks and Populations
• Activity driven modeling of time varying networks
• Novelty and Collective Attention
• Triadic closure as a basic generating mechanism of communities in complex networks
• Dynamical and bursty interactions in social networks
• Efficient local strategies for vaccination and network attack
• Social network dynamics of face-to-face interactions
• Improving immunization strategies
• An infectious disease model on empirical networks of human contact: bridging the gap between dynamic network data and contact matrices
• Least Cost Influence Maximization Across Multiple Social Networks
• Scale-free networks with an exponent less than two
• Three faces of node importance in network epidemiology: Exact results for small graphs