When the possible values of a target variable are specified and labeled, a model is considered supervised, that is, we know what we want to predict, and the goal is to find the most appropriate predictive model which will predict the outcome.
As an example, if we are predicting the approval rating for a product, we know what we are predicting (approval rating of a product), and we also usually know the range of possible outcomes. It could be a percentage from 0-100, or it could be some category such as high approval or low approval.
When the problem is supervised, the choice of which models to use are usually dependent upon the type of the target variable, that is, whether it is continuous or categorical. It can ...
