Logistic regression is a very common classification model. Strictly speaking, it is a classification model, and historical reasons are also called regression. Different from the derivation of parameters in linear regression, the method used here is no longer the least square method, but the maximum likelihood estimation. Logistic regression in the market is mostly used in spss, and its principle is rarely described.

Logistic regression can be regarded as a classification algorithm, but it can also be said to be an extension of linear regression. It is also classified as generalized linear regression because its derivation attempts to construct a linear model to explain the dependent variable with the idea of reduction. Now it is mostly used in medical statistics.