In supervised learning, the model learns from examples where the correct output is known. Training data consists of input-output pairs, and the model learns to map inputs to outputs. It powers classification (spam detection, image labeling) and regression (price prediction) tasks. It remains the most common and well-understood ML approach.









