silhouette
Confidence in Assignments via K-partition Ensembles
Exact and approximate silhouette scoring with micro, macro, and cluster-weighted averages.
Benchmark-first methods lane for Gnosis extraction in the Zer0pa portfolio: neutral route scoring, null metrics, stability battery, deterministic replay. Methodology surface; useful now, improving continuously.
A robust clustering evaluation framework that combines micro- and macro-averaged silhouette scores into a composite metric using statistical weighting.
