Unsupervised learning is a category of machine learning in which algorithms learn from data that does not have explicit labels or target outputs​

The goal is to discover underlying structure, groupings, or features in the data. Common unsupervised tasks include clustering (e.g., grouping customers by purchasing behavior), dimensionality reduction (e.g., compressing data with PCA while preserving variance), and density estimation. Unsupervised learning algorithms find patterns by leveraging the inherent similarities or differences between data points – for example, clustering algorithms will organize data into clusters based on a distance or similarity measure without any ground truth labels​

This type of learning is useful for exploratory data analysis, feature learning, anomaly detection, and as a preprocessing step for supervised tasks. Since no labeled data is required, unsupervised learning can make use of large amounts of raw data.