What is cross-validation in classification?

Cross-validation evaluates models using multiple train-test splits.

K-fold cross-validation splits data into K subsets, training on K-1 folds and testing on the remaining fold iteratively.