Editing Feature (machine learning) (section)

{{Short description|Measurable property or characteristic}}
{{distinguish|Feature (computer vision)}}
{{Refimprove|date=December 2014}}
{{machine learning bar}}
In [[machine learning]] and [[pattern recognition]], a '''feature''' is an individual measurable property or characteristic of a data set.<ref name="ml">{{cite book |author=Bishop, Christopher |title=Pattern recognition and machine learning |publisher=Springer |location=Berlin |year=2006 |isbn=0-387-31073-8 }}</ref> Choosing informative, discriminating, and independent features is crucial to produce effective [[algorithm]]s for [[pattern recognition]], [[Classification (machine learning)|classification]], and [[Regression analysis|regression]] tasks. Features are usually numeric, but other types such as [[String (computer science)|strings]] and [[Graph (discrete mathematics)|graphs]] are used in [[syntactic pattern recognition]], after some pre-processing step such as [[One-hot|one-hot encoding]]. The concept of "features" is related to that of [[Dependent and independent variables|explanatory variable]]s used in statistical techniques such as [[linear regression]].