Size Agnostic Change Point Detection Framework for Evolving Networks

TL;DR

This paper introduces a size-agnostic, fast framework for detecting change points in evolving networks that does not require prior knowledge of network size or historical data, demonstrating high accuracy on synthetic and real datasets.

Contribution

The authors propose a novel change point detection framework that is size agnostic and does not rely on historical or node identity information, improving detection performance.

Findings

Framework achieves high precision and recall

Outperforms previous change point detection methods

Effective on both synthetic and real-world datasets

Abstract

Changes in the structure of observed social and complex networks' structure can indicate a significant underlying change in an organization, or reflect the response of the network to an external event. Automatic detection of change points in evolving networks is rudimentary to the research and the understanding of the effect of such events on networks. Here we present an easy-to-implement and fast framework for change point detection in temporal evolving networks. Unlike previous approaches, our method is size agnostic, and does not require either prior knowledge about the network's size and structure, nor does it require obtaining historical information or nodal identities over time. We use both synthetic data derived from dynamic models and two real datasets: Enron email exchange and Ask-Ubuntu forum. Our framework succeeds with both precision and recall and outperforms previous solutions