Setting the Test Method
When the Start button is pressed, the selected learning algorithm is run and the
dataset that was loaded in the Preprocess panel is used with the selected test protocol.
For example, in the case of tenfold cross-validation this involves running the learning
algorithm 10 times to build and evaluate 10 classifiers. A model built from the
full training set is then printed into the Classifier Output area: This may involve
running the learning algorithm one final time. The remainder of the output depends
on the test protocol that was chosen using test options; these options were discussed
in Section 11.1.
Exercise 17.1.9. Load the iris data using the Preprocess panel. Evaluate C4.5
on this data using (a) the training set and (b) cross-validation. What is the
estimated percentage of correct classifications for (a) and (b)? Which estimate
is more realistic?