Sampling of Temporal Networks: Methods and Biases

TL;DR

This paper evaluates four sampling strategies for temporal networks, analyzing their biases on network statistics and epidemic modeling, and finds that uniform node sampling generally performs best across various scenarios.

Contribution

The study systematically compares sampling methods for temporal networks, highlighting biases and recommending best practices for data collection and analysis.

Findings

Uniform sampling of nodes reduces bias in most cases.

Sampling strategies significantly affect estimates of epidemic spread.

Some biases are consistent across different types of networks.

Abstract

Temporal networks have been increasingly used to model a diversity of systems that evolve in time; for example human contact structures over which dynamic processes such as epidemics take place. A fundamental aspect of real-life networks is that they are sampled within temporal and spatial frames. Furthermore, one might wish to subsample networks to reduce their size for better visualization or to perform computationally intensive simulations. The sampling method may affect the network structure and thus caution is necessary to generalize results based on samples. In this paper, we study four sampling strategies applied to a variety of real-life temporal networks. We quantify the biases generated by each sampling strategy on a number of relevant statistics such as link activity, temporal paths and epidemic spread. We find that some biases are common in a variety of networks and statistics, but one strategy, uniform sampling of nodes, shows improved performance in most scenarios. Our results help researchers to better design network data collection protocols and to understand the limitations of sampled temporal network data.