Health Data Science IV: Unsupervised Machine Learning in Health

4 Credits

Unsupervised machine learning is used to discover hidden patterns and structures in high-dimensional unlabeled health data. This course will survey leading techniques for clustering and dimensionality reduction. The course will cover hierarchical and density-based clustering techniques, along with modeling using Gaussian mixtures, factor analysis, and principal component analysis. Applications considered will include patient clustering for personalized treatment, anomaly detection for early disease identification, and dimensionality reduction for efficient analysis of diverse and complex medical datasets. Prerequisites: BSTA 141 and MATH 052 and MATH 043 and BSTA 040