mLFR Benchmark: Testing Community Detection Algorithms in Multi–layered, Multiplex and Multiple Social Networks
This is the extention of LFR Benchmark (https://sites.google.com/site/andrealancichinetti/files) introduced by A. Lancichinetti, S. Fortunato, F. Radicchi in the paper "Benchmark graphs for testing community detection algorithms". Community detection is one of the hottest topics in network science. While for simple (one layer) social networks there is hundreds of different algorithms for multilayer social networks there is a few. While for simple (one layered) social networks there is a number of reference dataset like karate club or football league, and few widely accepted and well tested benchmarks like GN Benchmark or LFR Benchmark for multilayer social networks there is a none. This paper propose an extension of well-known LFR Benchmark which will enable researchers to test and compare community detection algorithms in multilayer, multiplex and multiple social networks.
Bródka P. A Method for GroupExtraction and Analysis in Multi-layered Social Networks Ph.D. disertation, Wrocław, Poland, 2012 https://arxiv.org/abs/1612.02377
Bródka P., Grecki T.: mLFR Benchamark: Testing Community Detection Algorithms in Multilayer, Multiplex and Multiple Social Networks. https://github.com/pbrodka/mLFR-benchmark
If you are using our mLFR Benchmark please cite our work and work of A. Lancichinetti, S. Fortunato and F. Radicchi Lancichinetti, A., Fortunato, S., & Radicchi, F. (2008). Benchmark graphs for testing community detection algorithms. Physical review E, 78(4), 046110.
Repository contains
- Java application (jar)
- Java Source Code (Eclipse project)
PS. The code is from 2009-2010 so feel free to move it to Python, R or anything "newer" :)