This displays Accuracy measures for a classification model. Optionally, you can view the Confusion Matrix.
This data summarizes the information in a confusion matrix; this summary is particularly helpful when the target has more than two values. The following is displayed:
For more information about accuracy, see Accuracy Type.
To view the confusion matrix, click More Details at the bottom of the window. To hide the confusion matrix, click Less Detail.
By default, information about costs is not displayed. To see cost information and totals, click the Show Total and Cost checkbox.
The confusion matrix measures the likelihood of the model to predict incorrect and correct values; it also indicates the types of errors that the model is likely to make.
The confusion matrix is calculated by applying the model to a hold-out sample (the test set, created during the split step in a classification build activity) taken from the build data. The values of the target are known; the known values are compared with the values predicted by the model.
The columns are predicted values and the rows are actual values. For example, if you are predicting a target with values 0 and 1, the number in the upper right cell of the confusion matrix indicates the false-positive predictions, that is, predictions of 1 when the actual value is 0.
For more information about the confusion matrix, see Testing Classification and Regression Models.
Copyright © 2006, 2008, Oracle. All rights reserved.