Multiple-Particle-Walk

The Multiple Particle Competition and Cooperation or Multiple Particle Walk algorithm is a semi-supervised classification model.

It is a modified version of the original Particle Competition and Cooperation algorithm, see the reference for more information about the original.

The difference in this version is that each labeled sample will generate multiple particles instead of one and there is only random walk movement.

Installation

Install dependencies and the project in editable mode:

pip install -r requirements.txt
pip install -e .

Run example

python example.py

References:

[1] Breve, Fabricio Aparecido; Zhao, Liang; Quiles, Marcos Gonçalves; Pedrycz, Witold; Liu, Jiming, "Particle Competition and Cooperation in Networks for Semi-Supervised Learning," Knowledge and Data Engineering, IEEE Transactions on , vol.24, no.9, pp.1686,1698, Sept. 2012 doi: 10.1109/TKDE.2011.119