<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>http://en.zaoniao.it/index.php?action=history&amp;feed=atom&amp;title=Detection_error_tradeoff</id>
	<title>Detection error tradeoff - Revision history</title>
	<link rel="self" type="application/atom+xml" href="http://en.zaoniao.it/index.php?action=history&amp;feed=atom&amp;title=Detection_error_tradeoff"/>
	<link rel="alternate" type="text/html" href="http://en.zaoniao.it/index.php?title=Detection_error_tradeoff&amp;action=history"/>
	<updated>2026-05-15T21:30:34Z</updated>
	<subtitle>Revision history for this page on the wiki</subtitle>
	<generator>MediaWiki 1.32.0</generator>
	<entry>
		<id>http://en.zaoniao.it/index.php?title=Detection_error_tradeoff&amp;diff=4504&amp;oldid=prev</id>
		<title>Admin: Created page with &quot;Two hypothetical classifiers compared via DET curves. File:ext_38dKJsdjh_Example ROC curves.png|thumb|The same two cla...&quot;</title>
		<link rel="alternate" type="text/html" href="http://en.zaoniao.it/index.php?title=Detection_error_tradeoff&amp;diff=4504&amp;oldid=prev"/>
		<updated>2019-05-12T07:37:53Z</updated>

		<summary type="html">&lt;p&gt;Created page with &amp;quot;&lt;a href=&quot;/index.php?title=File:Ext_38dKJsdjh_Example_of_DET_curves.png&amp;amp;action=edit&amp;amp;redlink=1&quot; class=&quot;new&quot; title=&quot;File:Ext 38dKJsdjh Example of DET curves.png (page does not exist)&quot;&gt;thumb|Two hypothetical classifiers compared via DET curves.&lt;/a&gt; File:ext_38dKJsdjh_Example ROC curves.png|thumb|The same two cla...&amp;quot;&lt;/p&gt;
&lt;p&gt;&lt;b&gt;New page&lt;/b&gt;&lt;/p&gt;&lt;div&gt;[[File:ext_38dKJsdjh_Example of DET curves.png|thumb|Two hypothetical classifiers compared via DET curves.]]&lt;br /&gt;
[[File:ext_38dKJsdjh_Example ROC curves.png|thumb|The same two classifiers compared via traditional ROC curves.]]&lt;br /&gt;
A '''detection error tradeoff''' ('''DET''') '''graph''' is a graphical plot of error rates for [[binary classifier | binary classification]] systems, plotting the false rejection rate vs. false acceptance rate. The x- and y-axes are scaled non-linearly by their [[standard normal deviate]]s (or just by logarithmic transformation), yielding tradeoff curves that are more linear than [[Receiver operating characteristic |ROC curves]], and use most of the image area to highlight the differences of importance in the critical operating region.&lt;br /&gt;
&lt;br /&gt;
==Axis warping==&lt;br /&gt;
The normal deviate mapping (or normal quantile function, or inverse normal cumulative distribution) is given by the [[probit|probit function]], so that the horizontal axis is ''x'' = probit(''P&amp;amp;lt;sub&amp;amp;gt;fa&amp;amp;lt;/sub&amp;amp;gt;'') and the vertical is ''y'' = probit(''P&amp;amp;lt;sub&amp;amp;gt;fr&amp;amp;lt;/sub&amp;amp;gt;''),&lt;br /&gt;
where ''P&amp;amp;lt;sub&amp;amp;gt;fa&amp;amp;lt;/sub&amp;amp;gt;'' and ''P&amp;amp;lt;sub&amp;amp;gt;fr&amp;amp;lt;/sub&amp;amp;gt;'' are the false-accept and false-reject rates. &lt;br /&gt;
&lt;br /&gt;
The probit mapping maps probabilities from the unit interval [0,1], to the [[extended real line]] [−∞, +∞]. Since this makes the axes infinitely long, one has to confine the plot to some finite rectangle of interest.&lt;br /&gt;
&lt;br /&gt;
==See also==&lt;br /&gt;
&lt;br /&gt;
* [[Constant false alarm rate]]&lt;br /&gt;
* [[Detection theory]]&lt;br /&gt;
* [[False alarm]]&lt;br /&gt;
* [[Receiver operating characteristic]]&lt;br /&gt;
&lt;br /&gt;
==Source==&lt;br /&gt;
&lt;br /&gt;
[http://wikipedia.org/ http://wikipedia.org/]&lt;br /&gt;
[[Category:Error-correcting codes]]&lt;/div&gt;</summary>
		<author><name>Admin</name></author>
		
	</entry>
</feed>