The course objective is to teach how to use machine learning and statistical methods to identify clusters in multivariate data, i.e., groups of cases that have relatively high within-group similarity. Using those same methods, and additional ones, students will also learn how to identify cases that are relatively unique - anomalies (also called outliers). Students will first cover the building blocks - measuring distance between records and distance between clusters. Then students will learn how to use hierarchical clustering and k-means clustering algorithms, as well as normal mixture models to identify clusters (and, by extension, anomalies). Students will also learn some additional statistical methods for identifying anomalies.