In supervised learning, a user provides the machine with the desired input-output pairs as training data, that is, the data fed is into the machine as training data and is labeled. Whenever the machine encounters inputs similar to what it was trained with, it can define or classify the output according to the labeling it learned from the training data. Based on the labeled dataset, the machine would be able to find patterns in a new set of data.

One of the best examples of supervised learning is classification, where we can label a series of datasets as certain objects. Based on the labeled features and accuracy of the data being fed to the machine (training data), the machine would be able to classify newer datasets as those objects. Another example of supervised learning is regression, where, based on a training dataset and correlation between some parameters, the system can predict the output. Weather forecasting or detecting medical anomalies in human beings are scenarios where supervised learning is used.