My Rfuzzycoco package just hit the CRAN: https://cran.r-project.org/web/packages/Rfuzzycoco/index.html !! Publishing to CRAN is a rigorous process, and it was particularly challenging as this package includes custom C++ code. I documented the preparation process, including the steps needed for C++ integration, in a previous post: Preparing Rfuzzycoco for publication on CRAN

The Fuzzy CoCo Algorithm

The core algorithm, Fuzzy CoCo: a cooperative-coevolutionary approach to fuzzy modeling ingeniously combines fuzzy logic with cooperative genetic algorithms to evolve clear, human-understandable models, making it a powerful tool for explainable machine learning (XAI).

The C++ Foundation To make Rfuzzycoco possible, I first had to reimplement from scratch the main legacy Fuzzy CoCo implementation, which I released as the fuzzycoco C++ library, available here: https://github.com/Lonza-RND-Data-Science/fuzzycoco. You can read more details about this in my post: fuzzycoco: C++ open-source release of my re-implementation of the Fuzzy Coco algorithm.

Get started If you are interested in predicting or classifying your data with simple, human understandable, stable rules, give Rfuzzycoco a try, or reach out to me. I’m also open to collaborations, as there are many exciting opportunities to enhance both the implementation and the algorithm itself.

I (Karl Forner) am currently working as a consultant, contact me if you want me to help you with using R, organizing development, developing R packages or more generally supporting your software development efforts.