On Pruning and Averaging Decision Trees
Abstract
Pruning a decision tree is considered by some researchers to be the most important part of tree building in noisy domains. While, there are many approaches to pruning, an alternative approach of averaging over decision trees has not received as much attention. We perform an empirical comparison of pruning with the approach of averaging over decision trees. For this comparison we use a computationally efficient method of averaging, namely averaging over the extended, fanned set of a tree. Since there are a wide range of approaches to pruning, we compare tree averaging with a traditional pruning approach, along with an optimal pruning approach.
Cite
Text
Oliver and Hand. "On Pruning and Averaging Decision Trees." International Conference on Machine Learning, 1995. doi:10.1016/B978-1-55860-377-6.50060-8Markdown
[Oliver and Hand. "On Pruning and Averaging Decision Trees." International Conference on Machine Learning, 1995.](https://mlanthology.org/icml/1995/oliver1995icml-pruning/) doi:10.1016/B978-1-55860-377-6.50060-8BibTeX
@inproceedings{oliver1995icml-pruning,
title = {{On Pruning and Averaging Decision Trees}},
author = {Oliver, Jonathan J. and Hand, David J.},
booktitle = {International Conference on Machine Learning},
year = {1995},
pages = {430-437},
doi = {10.1016/B978-1-55860-377-6.50060-8},
url = {https://mlanthology.org/icml/1995/oliver1995icml-pruning/}
}