Centroid Decision Forest for High-Dimensional Classification
Description
Implements the Centroid Decision Forest (CDF) as a single user-facing
function CDF(). The method selects discriminative features via a multi-class
class separability score (CSS), splits by nearest class centroid, and aggregates
tree votes to produce predictions and class probabilities. Returns CSS-based
feature importance as well. Amjad Ali, Saeed Aldahmani, Zardad Khan (2025) .