Logistic regression

As described in the introduction, the logistic regression is indeed a classifier. It is possibly the most commonly used discriminative learning techniques for at least two reasons: its simplicity and the large variety of optimization algorithms used in the training of the model. Conceptually, logistic regression is the quantification of the relationship between an observed, target (or expected) variable, y, and a set of variables, x, that it depends on. Once the model is created (trained), it is available to classify real-time data.

Note

Generalized linear models

Logistic regression belongs to the broad category of generalized linear models (GLM). It relies on a linear combination of inputs or observations passed through a non-linear function, known as the logistic regression [9:13].

A logistic regression can be either binomial (two classes) or multinomial (three or more classes). In a binomial classification, the observed outcome is defined as {true, false}, {0, 1} or...