(n=8, d=20) User hybrid profiles per 48H episode

Clustering.py analyzes Profiling_Users_n_8_20_d.csv (assuming both files are in the same directory) to determine the optimal number of clusters between 1 and 8 using (elbow method). The upper bound is based on the rule-of-thumb k ≈ √(n/2) for n episodes. The algorithm clusters episodes using a hybrid feature space combining static socio-demographic attributes and dynamic software behavior metrics. The script outputs cluster assignments, validation metrics, and visualization artifacts for downstream analysis.