Internship← Artificial Intelligence & Machine Learning learning pathArtificial Intelligence & Machine Learning · Module 7 of 10
Unsupervised learning
Find structure in unlabelled data.
📖 Key ideas
- Clustering groups similar items: k-means, DBSCAN, hierarchical.
- Dimensionality reduction simplifies data: PCA, t-SNE, UMAP.
- Anomaly detection finds unusual records (fraud).
- Association rules find items bought together.
- Choose k with the elbow method or silhouette score.
💡 Real-life example: Clustering is a teacher grouping students by interests without being told the groups.
k-meansClustering into k groups
PCAPrincipal Component Analysis
Anomaly detectionFinding outliers
Elbow methodChoosing number of clusters