AAAI

Association for the Advancement of Artificial Intelligence

Home > Proceedings / Proceedings of the First International Conference on Knowledge Discovery and Data Mining > Book One

MDL-Based Decision Tree Pruning

Authors

Manish Mehta

Jorma Rissanen

and Rakesh Agrawal

IBM Almaden Research Center

Track:

All Contents

Downloads:

Download PDF

Abstract:

This paper explores the application of the Minimum Description Length principle for pruning decision trees. We present a new algorithm that more intuitively captures the primary goal of reducing the misclassification error. An experimental comparison is presented with three other pruning algorithms. The results show that the MDL pruning algorithm achieves good accuracy, small trees, and fast execution times.