Supervised Learning Tasks
Differing from unsupervised learning algorithms, supervised learning algorithms are characterized by their ability to find relationships between a set of features and a target value (be it discrete or continuous). Supervised learning can solve two types of tasks:
- Classification: The objective of these tasks is to approximate a function that maps a set of features to a discrete set of outcomes. These outcomes are commonly known as class labels or categories. Each observation in the dataset should have a class label associated with it to be able to train a model that is capable of predicting such an outcome for future data.
An example of a classification task is one that uses demographical data to determine someone's marital status.
- Regression: Although in regression tasks a function is also created to map a relationship between some inputs and some targets, in regression tasks, the outcome is continuous. This means that the outcome is a...
