What distinguishes unsupervised learning from supervised learning?

Unsupervised learning is characterized by the absence of labeled data, which is fundamental to its operations. In unsupervised learning, the algorithm is tasked with identifying patterns, structures, or relationships within the data without any external guidance or predefined labels. This allows the model to explore the inherent structure of the dataset through techniques such as clustering, association, or dimensionality reduction.

In contrast, supervised learning relies heavily on labeled data, where each training example includes both the input features and the corresponding output or label. This enables the model to learn the relationship between the input and output, making predictions on new data based on that learned mapping.

Recognizing this distinction is crucial for understanding how different machine learning methodologies apply to various types of data and application scenarios.