|
|
| Line 1: |
Line 1: |
| In [[statistical classification]], the '''Bayes error rate''' is the lowest possible error rate for a given class of classifier.<ref name=stat >Fukunaga, Keinosuke (1990) ''Introduction to Statistical Pattern Recognition'' by ISBN 0122698517 pages 3 and 97</ref><ref name=Tumer >K. Tumer, K. (1996) "Estimating the Bayes error rate through classifier combining" in ''Proceedings of the 13th International Conference on Pattern Recognition'', Volume 2, 695–699 </ref>
| | I'm Antoine and was born on 5 May 1970. My hobbies are Rock climbing and Agriculture Show.<br><br>Here is my website ... exterior home renovations ([http://www.homeimprovementdaily.com Read the Full Guide]) |
| | |
| A number of approaches to the estimation of the Bayes error rate exist. One method seeks to obtain analytical bounds which are inherently dependent on distribution parameters, and hence difficult to estimate. Another approach focuses on class densities, while yet another method combines and compares various classifiers.<ref name=Tumer />
| |
| | |
| The Bayes error rate finds important use in the study of patterns and [[machine learning]] techniques.{{cn|date=February 2013}}
| |
| | |
| ==Error determination==
| |
| In terms of machine learning and pattern classification, the data set can be discretely divided into 2 or more classes. Each element of the dataset is called an ''instance'' and the class it belongs to is called the ''label''.
| |
| The Bayes error rate of the dataset classifier is the probability of the classifier to incorrectly classify an instance.
| |
| For a [[multiclass classifier]], the Bayes error rate may be calculated as follows:{{cn|date=February 2013}}
| |
| | |
| :<math>p = \sum_{C_{i} \neq C_\text{max}} \textstyle \int\limits_{x\in H_{i}}P(x|C_{i})p(C_{i})\, dx,</math>
| |
| | |
| where ''x'' is an instance, ''C<sub>i</sub>'' is a class into which an instance is classified, ''H<sub>i</sub>'' is the area/region that a classifier function ''h'' classifies as ''C<sub>i</sub>''.{{clarify|reason= what is Cmax|date=February 2013}}
| |
| | |
| A Bayes error is non-zero if the distributions of the instances overlap, i.e. a certain instance ''x'' can have more than one label.{{cn|date=February 2013}}
| |
| | |
| ==See also==
| |
| * [[Naive Bayes classifier]]
| |
| | |
| ==References==
| |
| {{Reflist}}
| |
| | |
| [[Category:Statistical classification]]
| |
| [[Category:Bayesian statistics|Error rate]]
| |
| | |
| {{Statistics-stub}}
| |
I'm Antoine and was born on 5 May 1970. My hobbies are Rock climbing and Agriculture Show.
Here is my website ... exterior home renovations (Read the Full Guide)