Glossary · Machine learning
What is Cross-validation? Definition and meaning
A method that rotates which part of the data is held out for testing, averaging results across folds.
Definition
Cross-validation: A method that rotates which part of the data is held out for testing, averaging results across folds.
In short: Testing a model on data it has not seen.
When it is used
Model selection, hyperparameter tuning and honest performance estimates.
Where it fits
Cross-validation belongs to ML foundations in the Machine learning topic. These are the other terms in the same group:
- Linear regression: A model that predicts a numeric target as a weighted sum of input features, fitted by minimising squared error.
- Logistic regression: A linear model that outputs class probabilities through the logistic (sigmoid) function.
- Decision trees: A model that splits data step by step on feature thresholds, forming a tree of decisions that ends in predictions.
- Gradient boosting: An ensemble method that adds decision trees one at a time, each fitted to the remaining errors of the current model.
- k-nearest neighbors: A method that predicts by finding the k closest training examples and using their labels or values.
- Naive Bayes: A probabilistic classifier that applies Bayes' theorem assuming features are independent given the class.
- Principal component analysis: A linear method that finds orthogonal directions of maximum variance and projects data onto the top ones.
- t-SNE: A non-linear method that projects high-dimensional data to 2D or 3D, preserving local neighbourhoods for visualisation.
Learn more
The full guide to Cross-validation explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Cross-encoder reranking · Next term: CUDA