Unsupervised Learning

This section covers the main concepts and techniques in unsupervised learning.

Clustering

K-means Clustering

• Algorithm Overview
• Centroid Initialization
• Choosing K
• Evaluation Metrics

Hierarchical Clustering

• Agglomerative Clustering
• Divisive Clustering
• Linkage Methods
• Dendrogram Visualization

Density-Based Clustering

• DBSCAN
• OPTICS
• Mean-Shift
• Density Estimation

Advanced Clustering Methods

• Gaussian Mixture Models
• Spectral Clustering
• Affinity Propagation
• HDBSCAN

Dimensionality Reduction

Linear Methods

• Principal Component Analysis (PCA)
• Linear Discriminant Analysis (LDA)
• Factor Analysis
• Independent Component Analysis (ICA)

Non-linear Methods

• t-SNE
• UMAP
• Kernel PCA
• Locally Linear Embedding

Autoencoders

• Basic Autoencoders
• Denoising Autoencoders
• Variational Autoencoders
• Sparse Autoencoders

Association Rule Learning

Market Basket Analysis

• Support and Confidence
• Lift and Conviction
• Rule Generation
• Rule Evaluation

Algorithms

• Apriori Algorithm
• FP-Growth
• ECLAT
• Association Rule Mining

Anomaly Detection

Statistical Methods

• Z-score Method
• Modified Z-score
• Interquartile Range
• Statistical Tests

Machine Learning Methods

• Isolation Forest
• One-Class SVM
• Local Outlier Factor
• Autoencoders for Anomaly Detection

Evaluation and Validation

Clustering Evaluation

• Silhouette Score
• Davies-Bouldin Index
• Calinski-Harabasz Index
• Adjusted Rand Index

Dimensionality Reduction Evaluation

• Reconstruction Error
• Explained Variance
• Trustworthiness
• Continuity

Applications

Real-world Applications

• Customer Segmentation
• Document Clustering
• Image Compression
• Fraud Detection
• Recommendation Systems