GeneralConfusionMatrix Properties 
The GeneralConfusionMatrix type exposes the following members.
Name  Description  

Accuracy 
Accuracy. This is the same value as OverallAgreement.
 
ChanceAgreement 
Chance agreement.
 
ChiSquare 
Gets the ChiSquare statistic for the contingency table.
 
Classes 
Gets the number of classes.
 
ColumnErrors 
Gets the col errors.
 
ColumnProportions 
Gets the column marginals (proportions).
 
ColumnTotals 
Gets the column totals.
 
Cramer 
Cramer's V association measure.
 
Diagonal 
Gets the diagonal of the confusion matrix.
 
ExpectedValues 
Expected values, or values that could
have been generated just by chance.
 
GeometricAgreement 
Geometric agreement.
 
Kappa 
Gets the Kappa coefficient of performance.
 
Matrix 
Gets the confusion matrix, in which each element e_ij
represents the number of elements from class i classified
as belonging to class j.
 
Max 
Gets the maximum number of correct
matches (the maximum over the diagonal)
 
Min 
Gets the minimum number of correct
matches (the minimum over the diagonal)
 
OverallAgreement 
Overall agreement.
 
Pearson 
Pearson's contingency coefficient C.
 
PerClassMatrices 
Gets binary confusion matrices for each class in the multiclass
classification problem. You can use this property to obtain recall,
precision and other metrics for each of the classes.
 
Phi 
Phi coefficient.
 
Precision 
Gets the row precision.
 
ProportionMatrix 
Gets the confusion matrix in
terms of cell percentages.
 
Recall 
Gets the column recall.
 
RowErrors 
Gets the row errors.
 
RowProportions 
Gets the row marginals (proportions).
 
RowTotals 
Gets the row totals.
 
Sakoda 
Sakoda's contingency coefficient V.
 
Samples 
Gets the number of samples.
 
StandardError 
Gets the standard error of the Kappa
coefficient of performance.
 
StandardErrorUnderNull 
Gets the standard error of the Kappa
under the null hypothesis that the underlying Kappa
value is 0.
 
Tau 
Gets the Tau coefficient of performance.
 
TitleAboveColumns 
Gets or sets the title that ought be displayed on top of the columns of
this GeneralConfusionMatrix. Default is "Expected (Groundtruth)".
 
TitleOnTheLeftOfRows 
Gets or sets the title that ought be displayed on left side of this
GeneralConfusionMatrix. Default is "Actual (Prediction)".
 
Tschuprow 
Tschuprow's T association measure.
 
Variance 
Gets the variance of the Kappa
coefficient of performance.
 
VarianceUnderNull 
Gets the variance of the Kappa
under the null hypothesis that the underlying
Kappa value is 0.
