ClassificationSummary: Array;
ClassificationSummary: System.Array;
The ClassificationSummary property returns summary results of classification.
Classification results are presented as square matrix where number of rows and columns is equal to category number in IDataMining.Dependent and is equal to the number of elements in ISmDecisionTree.CategoriesList. Matrix rows correspond to real values of attribute for analysis, matrix columns correspond to forecasted values from ISmDecisionTree.FilledLearningSample.
Each element of the ClassificationSummary matrix is a number of objects from the source sample belonging to the category corresponding to the row, for those objects is forecasted that they belong to the category corresponding to the column where the element is placed.
Respectively the less elements are out of the ClassificationSummary main matrix diagonal, the better is classification quality and the better model can forecast object categories.
The property use is given in the example for ISmDecisionTree.Tree.
See also: