BLPKM-CC

Constrained clustering algorithm that considers must-link and cannot-link constraints.

Dependencies

BLPKM-CC depends on:

• Gurobi
• Numpy
• Scipy

Gurobi is a commercial mathematical programming solver. Free academic licenses are available here. A version of this algorithm that uses the non-commercial SCIP solver is available upon request. Please contact me by email (philipp.baumann@pqm.unibe.ch) if you are interested.

Installation

1. Download and install Gurobi (https://www.gurobi.com/downloads/)
2. Clone this repository (git clone https://github.com/phil85/BLPKM-CC.git)

Usage

The main.py file contains code that applies the BLPKM-CC algorithm on an illustrative example.

labels = blpkm_cc(X, n_clusters=2, ml=ml, cl=cl)

Documentation

The documentation of the module blpkm_cc can be found here.

Reference

Please cite the following paper if you use this algorithm.

Baumann, P. (2020): A Binary Linear Programming-Based K-Means Algorithm For Clustering with Must-Link and Cannot-Link Constraints. Proceedings of the 2020 IEEE International Conference on Industrial Engineering and Engineering Management, 324-328. → available online

Bibtex:

@inproceedings{baumann2020clustering,
	author={Philipp Baumann},
	booktitle={2020 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM)},
	title={A Binary Linear Programming-Based K-Means Algorithm For Clustering with Must-Link and Cannot-Link Constraints},
	year={2020},
	pages={324--328},
}

License

This project is licensed under the MIT License - see the LICENSE file for details