Logistic regression is a statistical approach used for binary classification problems in which the result variable is categorical and has two alternative outcomes, usually recorded as 0 and 1. Unlike linear regression, logistic regression predicts the likelihood of the output variable falling into a specific category based on the input characteristics. It uses the logistic function (also known as the sigmoid function) to convert the linear combination of input characteristics into a probability score ranging from 0 to 1. This makes logistic regression appropriate for estimating probabilities and categorizing observations into one of two groups depending on a predetermined threshold.
