Supervised learning is an ML approach used when labeled data is available to validate the model’s results. The core objective of models trained via supervised learning is to make predictions, specifically to split data into groups based on existing classes or labels.

Supervised tasks include Classification (where the output variable is discrete, producing categories or labels) and Regression (where the output variable is continuous, producing numbers). Named Entity Recognition (NER) and entity extraction are specifically categorized as supervised ML tasks.