Logistic regression
Logistic regression works by predicting class probabilities rather than actual classes. Ian Witten explains how this works, and how to use it.
Many classification methods produce probabilities rather than black-or-white classifications. Naive Bayes is an obvious example, but other methods do too. The numbers between 0 and 1 produced by linear regression in the previous lesson may look like probabilities, but they’re not. However, a variant called “logistic regression” does produce probabilities. Logistic regression is a powerful classification that predict probabilities directly through something called the “logit transform”.
