next up previous
Next: Training Set Size Up: EXPERIMENTS WITH BRUTEDL Previous: Default Rule Performance

Discussion

As illustrated by the results shown in the previous section, the default rule coverage decreases and the system's prediction accuracy increases as the size of the training set grows. However, the points at which the default rule's coverage starts to drop sharply and the system starts performing satisfactorily differ across domains. There are at least two subtle issues that influence these trends.





Jing Lin
Mon Apr 1 19:35:53 CST 1996