Which method is NOT typically used in unsupervised learning?

Unsupervised learning refers to the type of machine learning that deals with datasets without labeled outcomes. In this context, patterns, relationships, and structures in the data are discovered without the guidance of pre-existing labels.

Clustering is a primary method in unsupervised learning, as it involves grouping data points based on their similarities. Dimensionality reduction, another technique used in unsupervised learning, helps to reduce the number of features in a dataset while preserving essential information, making it easier to visualize and analyze data. Anomaly detection is also associated with unsupervised learning as it involves identifying outliers or unusual patterns in the data without prior labeling.

On the other hand, regression analysis falls under supervised learning, where the goal is to predict an outcome variable based on one or more predictor variables, using labeled data. Thus, it does not fit the paradigm of unsupervised learning, making it the method that is not typically associated with this category.