Exploration of scale-free networks

TL;DR

This paper investigates how measurement techniques, specifically tree-like explorations, can bias the observed properties of scale-free networks, raising concerns about data reliability.

Contribution

It demonstrates that certain exploration methods can alter the measured exponents of scale-free networks, highlighting potential biases in network data collection.

Findings

Tree-like exploration methods can change measured network exponents.

Measurement biases may affect the perceived properties of scale-free networks.

Data collection techniques influence the analysis of real network structures.

Abstract

The increased availability of data on real networks has favoured an explosion of activity in the elaboration of models able to reproduce both qualitatively and quantitatively the measured properties. What has been less explored is the reliability of the data, and whether the measurement technique biases them. Here we show that tree-like explorations (similar in principle to traceroute) can indeed change the measured exponents of a scale-free network.