The default classification model node builds several models using several different algorithms. If the target for the model is binary, four models are built, using all four of Oracle Data Mining's classification algorithms (Naive Bayes, Decision Tree, Support Vector Machine, and Generalized Linear Models). For information about these algorithms, see Classification Algorithms.
When you run the node, the input data set is split into build and test subsets by default.
Click the Classification icon in the Models section of the Component Palette.
Move the cursor to the workflow, and click again.
Change the name of the classification node to ClassBuild
by selecting the old name and typing the new one.
ClassBuild
has a warning indicator on the upper right corner. To see why it is not ready to run, float the cursor over the node:
Copyright © 2011, 2012, Oracle. All rights reserved.