Classification Model

In machine learning, classification is a predictive modeling process with the goal of determining which set of categories (class label) a given example of input data belongs to. The classifier uses some training data to understand how the given input variables relate to the class.

A classification model refers to supervised learning (algorithms which are designed to learn by example) and is represented by categorical data points or simple integers that represent each individual class.

For example, detecting spam at email providers is a binary classification problem as there are only two classes: spam and non-spam. In this case, known spam and non-spam messages should be used as training data. Once the classifier is accurately trained, it can be used to detect spam in unknown emails.