Python Multi-Objective Community Detection Algorithms
pymocd is a Python library, powered by a Rust backend, for performing efficient multi-objective evolutionary community detection in complex networks. This library is designed to deliver enhanced performance compared to traditional methods, making it particularly well-suited for analyzing large-scale graphs.
Navigate the Documentation for detailed guidance and usage instructions.
The HP-MOCD algorithm, central to pymocd, identifies community structures within a graph. It proposes a solution by grouping nodes into distinct communities, as illustrated below:
| Original Graph | Proposed Community Structure |
|---|---|
 |
 |
Installing the library using pip interface:
pip install pymocdFor an easy usage:
import networkx
import pymocd
G = networkx.Graph() # Your graph
alg = pymocd.HpMocd(G)
communities = alg.run()Important
Graphs must be provided in NetworkX or Igraph compatible format.
Refer to the official Documentation for detailed instructions and more usage examples.
We welcome contributions to pymocd! If you have ideas for new features, bug fixes, or other improvements, please feel free to open an issue or submit a pull request. This project is licensed under the GPL-3.0 or later.
If you use pymocd or the HP-MOCD algorithm in your research, please cite the following paper:
@article{santos2025high,
title={A High-Performance Evolutionary Multiobjective Community Detection Algorithm},
author={Santos, Guilherme O and Vieira, Lucas S and Rossetti, Giulio and Ferreira, Carlos HG and Moreira, Gladston},
journal={arXiv preprint arXiv:2506.01752},
year={2025}
}