Classification Matrix Tab (Mining Accuracy Chart View)

The Classification Matrix tab displays a classification matrix for each model selected in the models grid on the Column Mapping tab. The classification matrix is only available if the predictable column that is selected in the Column Mapping tab is not continuous. For a more detailed description of the Classification Matrix tab, see Validating Data Mining Models (Analysis Services - Data Mining).

Options

  • Copy
    Copies the content of each classification matrix to the clipboard.

  • Counts for <model> on < predictable column>
    Displays a classification matrix for each mining model in the mining structure. The matrix contains the following columns:

    Value

    Description

    Predicted

    Contains a row for each state of the predictable column.

    <states> (actual)

    A column for each state in the predictable column. If the state of the row and the column correspond, the cell represents the actual number of times the state exists in the database. If they do not correspond, the cell represents the error of the prediction.