Logistic Regression
Last Updated: July 29, 2026 | By Mihail Sebastian | AI Dictionary
A classification method, despite its name: it fits a sigmoid curve to data and outputs the probability that an input belongs to one of two classes.
What is Logistic Regression?
Logistic regression is a classification method that predicts the probability an input belongs to one of two classes – despite a name that suggests it predicts continuous values. The name survives from statistics, where the technique grew out of regression analysis; the output, though, is a probability between 0 and 1 that gets converted into a class label.
If you need to predict a number, use regression. If you need to predict a category, logistic regression is one of the first tools to reach for.
How Logistic Regression Works
The model computes a weighted sum of the input features, then passes it through the sigmoid function, which squashes any number into the range 0 to 1:
\[ p = \sigma(w \cdot x + b), \qquad \sigma(z) = \frac{1}{1 + e^{-z}} \]Training adjusts the weights until the predicted probabilities match the labels in the training data. A threshold, usually 0.5, then turns each probability into a decision: above it, one class; below it, the other.
Each weight has a readable meaning: it states how strongly its feature pushes the odds up or down. That is why logistic regression persists in regulated fields, where a lender or hospital must explain every automated decision.
Example of Logistic Regression
A bank builds a default-risk model from past loans. Features include income, existing debt, and payment history; the label records whether each borrower defaulted.
Training finds that a high debt-to-income ratio raises the weighted sum and past late payments raise it further, while a long clean record lowers it. A new applicant’s numbers produce a sum of 1.2, and the sigmoid turns that into a default probability of 0.77.
The bank auto-approves only below a 0.2 default probability, so this application routes to a human reviewer. Because every weight is visible, the bank states exactly which factors drove the score when fair-lending rules require an explanation.
Related AI terms: Classification · Sigmoid Function · Regression Analysis · Naive Bayes
Did you like the Logistic Regression gist?
Learn about 250+ need-to-know artificial intelligence terms in the AI Dictionary.
Mihail Sebastian — Writes about AI governance, regulation, and the technology behind them. Placeholder bio — replace with a real credential line. About