Glossary

Classification

A supervised learning task that assigns each input to one of a fixed set of discrete categories or labels.

Also called: classifier, classification model

Classification is a supervised learning task in which a model learns to assign each input record to one of a fixed set of discrete categories, such as "fraud" or "not fraud," or one of several product types. It differs from regression, which predicts a continuous number instead of a label, though many of the same algorithms are adapted for both.

A classification model is trained on examples where the correct label is already known, and learns a decision boundary that separates classes based on input features. Common approaches include logistic regression, decision tree and random forest models, and neural networks. Output is typically a predicted class plus a probability or confidence score, which is turned into a hard label by applying a threshold.

Classification underlies tasks such as churn prediction, spam filtering, credit approval and medical diagnosis support. Practitioners evaluate it with a confusion matrix and metrics derived from it rather than plain accuracy, because accuracy is misleading when classes are imbalanced, for example when fraud is rare. A model that always predicts "not fraud" can score over 99% accuracy while catching no actual fraud, which is why precision, recall and related measures are the standard way to judge a classifier's real usefulness.

Last reviewed September 22, 2026

In the index now

Related terms

Supervised learning A machine learning approach where a model learns from examples that already include the correct answer, to predict that answer on new data. Regression analysis A statistical method for estimating the relationship between one outcome variable and one or more predictor variables. Decision tree A supervised model that predicts an outcome by following a sequence of if-then splits on feature values, shaped like a tree. Random forest An ensemble model that trains many decision trees on random subsets of data and features, then averages their predictions. Confusion matrix A table comparing a classifier's predicted labels against actual labels, breaking results into true and false positives and negatives.

Related guides

How to analyze text data From word counts and topic models to LLM classification — matching the method to the question instead of reaching for the newest tool by default.