next up previous
Next: Tradeoff Simplicity-Accuracy Up: Inducing Interpretable Voting Classifiers Previous: Optimistic Pruning

Experiments

Following are three experimental sections, aimed at testing WIDC on three issues. The first presents extensive results on the tradeoff simplicity-accuracy obtained by WIDC, and compares the results with those obtained for state-of-the-art algorithms. The second goes on in depth analyzes for the mining/interpretability issue, and the third presents results on noise tolerance.



Subsections

2002 AI Access Foundation and Morgan Kaufmann Publishers. All rights reserved.