AI Classification Explainer
Classification is a machine-learning task that maps an input to a discrete category. A model might label an email as spam or not spam, identify the subject of an image, sort a support request, or estimate whether a transaction belongs to a risk class.
During supervised training, examples pair features with labels. The model learns parameters that separate or score the classes. At inference time it often produces probabilities or scores, and a decision rule chooses one class, several labels, or an option to abstain.
Binary classification has two outcomes, multiclass classification chooses among more than two mutually exclusive classes, and multilabel classification allows several labels at once. The correct formulation depends on what the labels mean and how ambiguous cases should be handled.
Accuracy can be misleading when one class is rare or errors have unequal consequences. Precision, recall, false-positive rate, confusion matrices, calibration, subgroup performance, and threshold tradeoffs provide a fuller view. Labels and test data must also reflect the real deployment setting.
Classification is a machine-learning task that assigns an input to one or more discrete categories based on patterns learned from examples.
Key terms include feature, label, class, score, probability, threshold, binary, multiclass, multilabel and confusion matrix.
Many classifiers output one score per class. A threshold or maximum-score rule converts those values into a decision, while an abstention rule can defer uncertain cases.
Changing the threshold shifts the balance between false positives and false negatives without retraining the underlying model.
A high overall accuracy may hide poor detection of the rare class or much worse performance for a subgroup that appears infrequently in the data.
Evaluation should connect metrics to consequences, include representative examples, check calibration, and preserve human review for consequential or ambiguous cases.
Explore more "Explainers"
Discover additional explainers across politics, science, business, technology, and other fields. Each explainer breaks down a complex idea into clear, everyday language—helping you better understand how major concepts, systems, and debates shape the world around us.