Which machine learning approach focuses on discovering patterns in data without predefined labels?

Unsupervised learning is a machine learning approach that focuses on discovering patterns in data without the use of predefined labels. This means that the algorithm is tasked with exploring the input data to identify intrinsic structures, groupings, or relationships that may exist among the data points.

In contrast to supervised learning, where the model is trained on labeled data (i.e., each training example comes with a label indicating the correct output), unsupervised learning relies solely on the input data to find hidden patterns or clusters. Common techniques in unsupervised learning include clustering (like K-means or hierarchical clustering) and dimensionality reduction (like PCA, principal component analysis).

This approach is particularly useful in scenarios where labeled data is scarce or expensive to obtain, as it allows for analysis of large datasets without the need for extensive labeling efforts. Examples of practical applications include market segmentation, anomaly detection, and exploratory data analysis, where understanding the underlying structure of the data is essential.