Multilayer Artificial Benchmark for Community Detection (mABCD)

2025-07-14

AI Summary

Existing multilayer network community detection lacks interpretable and analytically tractable benchmark models. Method: We propose Multi-ABCD—the first ABCD model variant specifically designed for multilayer networks—built upon the hierarchical configuration model framework. It explicitly encodes cross-layer node memberships and intra-/inter-layer connection mechanisms, enabling generation of multilayer random graphs with power-law degree distributions, power-law community-size distributions, and controllable community structure. Inter-layer noise parameters and an analytical construction strategy jointly ensure theoretical tractability, computational efficiency, and fidelity to real-world network properties. Contribution/Results: Experiments demonstrate that Multi-ABCD-generated LFR-like multilayer benchmarks surpass existing approaches in scalability and statistical fidelity. The model thus provides a theoretically grounded, practical benchmarking tool for systematic evaluation of multilayer community detection algorithms.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Machine Learning: Multi-instance/Multi-view Learning
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Models for Web evolution

Abstract

The Artificial Benchmark for Community Detection (ABCD) model is a random graph model with community structure and power-law distribution for both degrees and community sizes. The model generates graphs similar to the well-known LFR model but it is faster, more interpretable, and can be investigated analytically. In this paper, we use the underlying ingredients of the ABCD model and introduce its variant for multilayer networks, mABCD.

Problem

• Extends ABCD model to multilayer networks
• Generates graphs with community structure
• Improves speed and interpretability over LFR

Innovation

• Extends ABCD model to multilayer networks
• Maintains power-law degree distribution
• Ensures fast and interpretable community detection