Modeling and Characterizing Complex Contagion Processes on Time-Varying Networks
Abstract
Social norms, (mis)information, beliefs and innovations diffuse through social networks via complex contagion processes. These phenomena are very different from biological contagions. In fact, complex (social) contagions are active rather than passive acts, might be advantageous, require multiple exposures, and are linked to social legitimation, credibility and uncertainty among other factors. The literature devoted to the study of complex contagions is vast yet affected by two major limitations: it largely neglects 1) the time-varying nature of the social networks where the spreading takes place 2) the co-evolution between social contagions and networksÕ dynamics. By using time-varying networks modeling and data science approaches the proposed research aims at filling these gaps. In particular, the proposal is organized around two main tasks. The first will focus on the study of complex contagion processes unfolding on artificial time-varying networks. By using the activity-driven modeling framework for temporal graphs the project will explore both analytically and numerically how major time-varying features of real networks affect social contagions. Examples of such properties are burstiness, local and global correlations in activations, as well as the presence of communities. The activity-driven framework is modular, thus the project will be able to study the impact of such features both separately and in combination. The second task instead will be devoted to the development of adaptive models of time-varying networks where the contagion processes and social interactions co-evolve. In doing so, the proposal will study the feedback loop between the dynamics on and of networks considering observations and insights from real online social networks and collaborations systems. In this context, the project will explore the efficiency of different mechanisms aimed at promoting interactions across otherwise separated groups clustered around different opinions, norms, and beliefs. Overall, the proposal will advance our understanding of social contagion processes unfolding on social networks both from a theoretical and a practical standpoint. In doing so, the project will pave the way to address pressing societal issues such as the polarization and the emergence of filter bubbles in modern interactions.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 14, 2019
- Source ID
- W911NF1810376
Entities
People
- Nicola Perra
Organizations
- Army Contracting Command
- United States Army
- University of Greenwich