CCADMACL - Advanced Machine Learning

3rd Year • 2nd Term

This repository contains lecture notes, notebooks, and exercises for the Advanced Machine Learning course.
The primary focus is on unsupervised learning, probabilistic models, density estimation, and pattern mining.

---

Course Outline

Week 2 — Introduction to Unsupervised Machine Learning

• Supervised vs. Unsupervised vs. Semi-supervised learning
• Applications:
  • Clustering
  • Dimensionality reduction
  • Anomaly detection
  • Topic modeling
• Types of unsupervised problems:
  • Pattern discovery
  • Density estimation
• Generative vs. Discriminative models

---

Week 3–4 — Introduction to Clustering

• Cluster data to group similar points without labels
• Distance metrics:
  • Euclidean
  • Manhattan
  • Cosine
  • Mahalanobis
• K-Means Algorithm:
  • Intuition
  • Initialization (k-means++)
  • Convergence
• Hierarchical Clustering:
  • Agglomerative
  • Divisive
• Density-based clustering:
  • DBSCAN
  • OPTICS
• Evaluation metrics:
  • Silhouette score
  • Davies–Bouldin index
  • Adjusted Rand index

---

Week 5-6 — Introduction to Dimensionality Reduction

• Curse of dimensionality
• Principal Component Analysis (PCA):
  • Covariance matrix
  • Eigenvectors
  • Variance explained
• Singular Value Decomposition (SVD)
• t-SNE and UMAP for nonlinear embedding
• Feature selection vs. feature extraction

---

Week 7-8 — Probabilistic Models & Density Estimation

• Gaussian Mixture Models (GMM)
• Expectation-Maximization (EM) algorithm

---

Week 9-10 — Clustering Evaluation and Validation

• Measure and interpret unsupervised model performance
• Metrics:
  • Silhouette score
  • Inertia
  • Purity
  • Entropy
• Visualization techniques:
  • Elbow method
  • t-SNE projection

---

Week 11-12 — Association Rule Learning

• Market Basket Analysis
• Apriori Algorithm
• Eclat Algorithm
• Frequent Pattern Growth Algorithm (FP-Growth)
• Metrics:
  • Support
  • Confidence
  • Lift