multi-class classification

Multi-class classification is a supervised learning task in which a model assigns each example to one of three or more mutually exclusive classes. For instance, an image classifier might choose whether a picture shows a cat, a dog, or a bird. A model can produce a score or probability for each class, and a decision rule—often selecting the highest-scoring class—turns those outputs into a prediction. Multi-class problems differ from binary classification, which has two classes, and from multi-label classification, where several labels may apply to the same example. Evaluation may use accuracy, a confusion matrix, or per-class precision and recall, especially when classes are imbalanced or the costs of different mistakes vary.