This C++ project implements and evaluates the performance of K-Means clustering using internal validation metrics such as the Calinski-Harabasz (CH) Index and Davies-Bouldin (DB) Score. It also includes preprocessing steps like data normalization and supports running multiple randomized experiments to find optimal cluster configurations.

Code repository for the manuscript: Violent victimisation after adolescent traumatic brain injury: development and validation of a clinical prediction model (under review)

CompSil: Composite Silhouette for Cluster-Count Selection