A type of machine learning where an algorithm learns to generate predictions using labeled training data.
Supervision in machine learning refers to the availability of labeled training data. Each data point is connected with its intended outcome. This supervision acts as a guide for the algorithm, helping to alter its internal parameters.
There are two types of supervised learning tasks that are commonly used with machine learning algorithms: Classification and regression.
Data is partially labelled, we have many pairs (X,y), we may also have many Xs without known ys. We would like to estimate a function f() so that y=f(X).
When y is a real number, it is a regression problem,
When y is class label (or categorical), it is classification problem.