Definition
Holdout evaluation is an approach to out-of-sample evaluation whereby the available data are partitioned into a training set and a test set. The test set is thus out-of-sample data and is sometimes called the holdout set or holdout data. The purpose of holdout evaluation is to test a model on different data to that from which it is learned. This provides less biased estimate of learning performance than in-sample evaluation.
In repeated holdout evaluation, repeated holdout evaluation experiments are performed, each time with a different partition of the data, to create a distribution of training and test sets with which an algorithm is assessed.
Cross-References
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer Science+Business Media New York
About this entry
Cite this entry
(2017). Holdout Evaluation. In: Sammut, C., Webb, G.I. (eds) Encyclopedia of Machine Learning and Data Mining. Springer, Boston, MA. https://doi.org/10.1007/978-1-4899-7687-1_369
Download citation
DOI: https://doi.org/10.1007/978-1-4899-7687-1_369
Published:
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4899-7685-7
Online ISBN: 978-1-4899-7687-1
eBook Packages: Computer ScienceReference Module Computer Science and Engineering