Lesson 01
Lloyd's algorithm, K-means++; chọn k bằng elbow/silhouette; agglomerative hierarchical, dendrogram.
Lesson 02
Density-reachable, eps + minPts; vs K-means cho non-convex (moons, rings); outlier detection; HDBSCAN brief.
Lesson 03
PCA qua eigendecomposition covariance / SVD; variance explained, scree plot — bridge Vectors/04 Linear Algebra.
Lesson 04
t-SNE (perplexity, KL divergence); UMAP (faster, preserves global); pitfalls (cluster size, distance interpretation).