Using Networks Analysis and Visualization to Explain COVID-19 Spread through the Physical, Social, and Information Graphs

Presentation Date: 

Thursday, June 9, 2022

Location: 

New York, NY

2022 C+J Conference at Columbia Journalism School
https://cj2022.brown.columbia.edu/schedule
Register

Understanding network science and compartmental models in epidemiology are essential to understanding, explaining, intervening, and forecasting the pandemic. This workshop introduces core concepts and skills for network data analysis and visualization using Flourish and NetworkX. We begin with simple data sets that represent the flow of migration between countries using sankey, chord, correlation matrix heatmaps and, thereby, introduce the notion of nodes and edges in a graph. Next, we introduce network models such as small-world and scale-free networks to describe typical properties of complex networks, and demonstrate how to use Flourish to produce network visualizations, encoding the size of nodes as centrality and thickness of the links as edge weight. Then, we use NetworkX to analyze network statistics and run a simple agent-based model simulation of misinformation diffusion based on a simplified SIR epidemic model. By the end of the workshop, participants will fully appreciate the network science and design principles that inspired the the fam visual story created by Harry Stevens for the Washington Post: 'Why outbreaks like coronavirus spread exponentially, and how to “flatten the curve”'. We end with a brainstorming discussion of ways to combine interaction of three different layers in the multiplex network--physical mobility, social relationship, and infodemic beliefs--drive outcomes and trajectories for the ongoing pandemic.