Artificial Benchmark for Community Detection with Outliers (ABCD+o)

TL;DR

This paper introduces an extended artificial benchmark model, ABCD+o, for community detection that incorporates outliers, enabling more realistic testing of algorithms on graphs with community structures and outliers.

Contribution

The paper extends the ABCD model to include outliers, providing a more realistic benchmark for community detection algorithms.

Findings

Outliers in ABCD+o have distinguishable properties.

The model can mimic real-world network outliers.

Experimental results validate the model's effectiveness.

Abstract

The Artificial Benchmark for Community Detection graph (ABCD) is a random graph model with community structure and power-law distribution for both degrees and community sizes. The model generates graphs with similar properties as the well-known LFR one, and its main parameter $\xi$ can be tuned to mimic its counterpart in the LFR model, the mixing parameter $\mu$. In this paper, we extend the ABCD model to include potential outliers. We perform some exploratory experiments on both the new ABCD+o model as well as a real-world network to show that outliers possess some desired, distinguishable properties.