Dynamical Properties of Interaction Data

TL;DR

This paper introduces a novel method for analyzing network dynamics by creating a 'temporal web' that encodes interactions over time, enabling static network measures to reveal features of transmission dynamics in epidemiological models.

Contribution

It presents a new approach to study network dynamics through a temporally extended network representation called 'temporal web,' applied to epidemiological models.

Findings

The 'temporal web' captures transmission dynamics effectively.

Static network measures applied to the temporal web reveal key features.

Method successfully applied to SEIR and SEIS models.

Abstract

Network dynamics are typically presented as a time series of network properties captured at each period. The current approach examines the dynamical properties of transmission via novel measures on an integrated, temporally extended network representation of interaction data across time. Because it encodes time and interactions as network connections, static network measures can be applied to this "temporal web" to reveal features of the dynamics themselves. Here we provide the technical details and apply it to agent-based implementations of the well-known SEIR and SEIS epidemiological models.