Synthetic Modular Graphs from LFR and ABCD Benchmark Models for Community Detection

50 randomly generated network files from a study on community detection algorithms. The dataset includes 30 ABCD graphs and 20 LFR graphs, created by Samin Aref and Mahdi Mostajabdaveh for their 2024 Journal of Computational Science article. Each network is provided in .gml format under a CC BY-NC-SA 4.0 license.

Use Cases

• Benchmarking community detection algorithms based on synthetic modular graphs.
• Analyzing the performance of graph neural networks on structured benchmark networks.
• Studying the properties of LFR and ABCD graph generation models for network science.

Strengths

• Contains 50 distinct network files, providing a substantial test suite.
• Includes graphs from two established benchmark models: 30 ABCD and 20 LFR graphs.
• Explicitly linked to a peer-reviewed 2024 research article for methodological context.

Limitations

• Column-level documentation is absent; node and edge attribute semantics must be inferred after download.
• Row count and overall graph scale (nodes/edges) are unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Samin Aref via Figshare (DOI: 10.6084/m9.figshare.24257293).

Collection Method

Randomly generated using LFR and ABCD benchmark graph models.