DD_tools: dd_auc

 DD_AUC compute the integrated error under the ROC curve

   ERR = DD_AUC(E,BND)

 Compute the integrated error under the ROC curve, given in E (as
 obtained by dd_roc).
 The user can supply a lower and upper bound on the FP (outliers
 accepted) over which the ROC is integrated: BND = [lowerbnd,
 upperbnd], for instance BND = [0.05 0.5].

 IMPORTANT: According to the definition of AUC, high values are good
 (so it is a performance measure, not an error measure). Therefore
 the definition has been changed!! Take care!!

   ERR = A*W*DD_AUC

 A new format is to apply DD_AUC as a mapping, by multiplying the
 classifier output with DD_AUC.

See also: dd_error dd_roc 

@article{Bradley1997,
	author = {Bradley, A.P.},
	title = {The use of the area under the {ROC} curve in the
	evaluation of machine learning algorithms},
	journal = {Pattern Recognition}, year = {1997}, volume = {30},
	number = {7}, pages = {1145-1159}
}



Copyright: D.M.J. Tax, D.M.J.Tax@prtools.org
Faculty EWI, Delft University of Technology
P.O. Box 5031, 2600 GA Delft, The Netherlands