Unsupervised Machine Learning

This module introduces unsupervised learning concepts and K-means clustering. Students learn clustering theory, K-means algorithm implementation, initialization methods, and techniques for selecting optimal cluster numbers.

This module covers various distance metrics used in clustering, including Euclidean, Manhattan, Cosine, and Jaccard distances. It also addresses the curse of dimensionality and its impact on clustering.

This module explores different clustering algorithms including Hierarchical Agglomerative Clustering, DBSCAN, and Mean Shift. Students learn to compare algorithms and select appropriate techniques for different data types.

This module introduces dimensionality reduction techniques with a focus on Principal Component Analysis (PCA). It covers applications in big data, imaging, and data preprocessing.

This module covers advanced dimensionality reduction techniques including Kernel PCA and multidimensional scaling, offering more powerful alternatives to standard PCA.

This module explores matrix factorization techniques, particularly Non-Negative Matrix Factorization, for applications in big data, text mining, and data preprocessing.

This module consists of a final project where students apply their unsupervised learning knowledge to demonstrate their skills.