Courses / HEC NSCT – National Skill Competency Test: Complete Preparation for Computing Graduates
8.4 Unsupervised Learning
Area 8 – AI, Machine Learning & Data Analytics (10%) · 35 min read · 6 MCQs to check yourself
Unsupervised learning finds structure in unlabelled data.
| Task | Algorithms | Example |
|---|---|---|
| Clustering | k-means (choose k; assign to nearest centroid; recompute; repeat), hierarchical (dendrogram), DBSCAN (density, finds outliers) | Customer segmentation |
| Dimensionality reduction | PCA, t-SNE | Compress features, visualise high-dimensional data |
| Association rules | Apriori (support, confidence, lift) | Market basket: bread → butter |
| Anomaly detection | Isolation forest | Fraud detection |
- Choose k for k-means with the elbow method or silhouette score.
- Semi-supervised: few labels + many unlabelled examples. Reinforcement learning: an agent learns by rewards and penalties (game playing, robotics) – Q-learning.
K-means clustering (StatQuest)
This lesson ends with a quiz of 6 MCQs. Open the lesson on PrepOasis to take it and save your progress.
Previous: 8.3 Supervised Learning and Model Evaluation · Next: 8.5 Deep Learning, NLP, Computer Vision and Responsible AI