Can you explain the difference between supervised and unsupervised learning?

Example Answer

The way I'd explain it in an interview is this: Supervised learning uses labeled examples, meaning the model sees inputs and the target output it is supposed to predict. The goal is usually prediction: classification, regression, ranking, or some other task where the desired outcome is known during training.

Unsupervised learning works without explicit target labels. Instead, the model tries to uncover structure in the data, such as clusters, lower-dimensional structure, anomalies, or latent patterns. In practice, the difference is not just whether labels exist. It is whether the learning objective is tied to a known prediction target or to discovering structure on its own. That difference shapes how you evaluate the model, how you collect data, and what kinds of business problems each approach can solve well.

Common Poor Answer

A weak answer says supervised learning uses labeled data and unsupervised learning does not, without explaining the difference in objective, use case, and evaluation.