Synonyms
Definition
The cost or loss of a prediction y′, when the correct value is y, is a measure of the relative utility of that prediction given that correct value. A common loss function used with classification learning is zero-one loss. Zero-one loss assigns 0 to loss for a correct classification and 1 for an incorrect classification. Cost sensitive classification assigns different costs to different forms of misclassification. For example, misdiagnosing a patient as having appendicitis when he or she does not might be of lower cost than misdiagnosing the patient as not having it when he or she does. A common loss function used with regression is error squared. This is the square of the difference between the predicted and true values.
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). Loss. 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_499
Download citation
DOI: https://doi.org/10.1007/978-1-4899-7687-1_499
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