RubixML / ML

A high-level machine learning and deep learning library for the PHP language.
https://rubixml.com
MIT License
2.03k stars 182 forks source link

Evaluation of the cluster quality with indicators #304

Open cammarano opened 1 year ago

cammarano commented 1 year ago

Supposing to start from an unlabeled dataset featured by N features, I want to perform a clustering. I don't have a testing sample (labels) and ground truth is not available. I don't see a tool for evaluating the clustering quality, for instance to provide silhouette coefficients. Is there a way to perform this kind of analysis?

andrewdalpino commented 1 year ago

Hey @cammarano, yeah so far all the clustering metrics and reports require the ground-truth. I would love to implement something like a silhouette score some day.