Cross-Validation
A way to test a model by splitting the data into several folds, training on most and testing on the rest, then rotating so every fold gets a turn as the test set. You get a sturdier estimate than a single split.
Think of It Like This
Grading with several different pop quizzes instead of betting everything on one big exam.
In k-fold cross-validation you slice the data into k parts, train k times — each run leaving one part out to test on — and average the scores. That uses your data efficiently and reveals whether a result is stable or just a fluke of one lucky split. It's also how you compare models and tune hyperparameters without peeking at the final test set, which makes it a strong guard against overfitting.