Maximum Likelihood Rule Ensembles
Abstract
We propose a new rule induction algorithm for solving classification problems via probability estimation. The main advantage of decision rules is their simplicity and good interpretability. While the early approaches to rule induction were based on sequential covering, we follow an approach in which a single decision rule is treated as a base classifier in an ensemble. The ensemble is built by greedily minimizing the negative loglikelihood which results in estimating the class conditional probability distribution. The introduced approach is compared with other decision rule induction algorithms such as SLIPPER, LRI and RuleFit.
Cite
Text
Dembczynski et al. "Maximum Likelihood Rule Ensembles." International Conference on Machine Learning, 2008. doi:10.1145/1390156.1390185Markdown
[Dembczynski et al. "Maximum Likelihood Rule Ensembles." International Conference on Machine Learning, 2008.](https://mlanthology.org/icml/2008/dembczynski2008icml-maximum/) doi:10.1145/1390156.1390185BibTeX
@inproceedings{dembczynski2008icml-maximum,
title = {{Maximum Likelihood Rule Ensembles}},
author = {Dembczynski, Krzysztof and Kotlowski, Wojciech and Slowinski, Roman},
booktitle = {International Conference on Machine Learning},
year = {2008},
pages = {224-231},
doi = {10.1145/1390156.1390185},
url = {https://mlanthology.org/icml/2008/dembczynski2008icml-maximum/}
}