resubEdge

Class:ClassificationDiscriminant

Classification edge by resubstitution

expand all in page

Syntax

edge = resubEdge(obj)

Description

edge= resubEdge(obj)returns the classification edge obtained byobjon its training data.

Input Arguments

obj

Discriminant analysis classifier, produced usingfitcdiscr.

Output Arguments

edge

Classification edge obtained by resubstituting the training data into the calculation of edge.

Examples

expand all

Estimate the Resubstitution Edge of Discriminant Analysis Classifiers

Open Live Script

Estimate the quality of a discriminant analysis classifier for Fisher's iris data by resubstitution.

Load Fisher's iris data set.

loadfisheriris

Train a discriminant analysis classifier.

Mdl = fitcdiscr(meas,species);

Compute the resubstitution edge.

redge = resubEdge(Mdl)

redge = 0.9454

More About

expand all

Edge

Theedgeis the weighted mean value of the classificationmargin. The weights are class prior probabilities. If you supply additional weights, those weights are normalized to sum to the prior probabilities in the respective classes, and are then used to compute the weighted average.

Margin

The classificationmarginis the difference between the classificationscorefor the true class and maximal classification score for the false classes.

The classification margin is a column vector with the same number of rows as in the matrixX. A high value of margin indicates a more reliable prediction than a low value.

Score

For discriminant analysis, thescoreof a classification is the posterior probability of the classification. For the definition of posterior probability in discriminant analysis, seePosterior Probability.