Which of the following best describes the role of supervised learning?

Supervised learning is a machine learning paradigm where models are trained on labeled datasets. This means that the data used for training includes both the input features and the corresponding output labels. The primary objective of supervised learning is to learn a mapping from inputs to outputs so that the model can make accurate predictions when provided with new, unseen data.

In this context, using labeled datasets enables the model to understand the relationship between features and the target outcome during the training process. As a result, supervised learning is particularly effective for classification and regression tasks, allowing practitioners to build applications that can predict categories or numerical values based on the patterns learned from the data.

The other options describe various aspects of machine learning and data analysis but do not specifically characterize the role of supervised learning. For example, discovering patterns without labeled data is indicative of unsupervised learning, while applying random transformations to data often relates to data augmentation techniques. Establishing correlations between different datasets refers to methods like correlation analysis, which is not exclusive to supervised learning. Thus, the chosen answer accurately encapsulates the essence of supervised learning.