Discovering Production Rules with Higher Order Neural Networks
Abstract
The paper demonstrates by example that neural networks can be used successfully for automatic extraction of production rules from empirical data. The particular case considered here is a popular public domain data base of 8,124 mushrooms. With the use of a term selection algorithm described in the paper we developed a number of very accurate high order networks (polynomial classifiers). Then rounding of synaptic weights was applied leading in many cases to networks with integer weights which were subsequently converted to production rules. We also show that “focusing of network attention― to a smaller subset of useful attributes ordered with respect to their decreasing discriminating abilities helps significantly in accurate rule generation.
Cite
Text
Kowalczyk et al. "Discovering Production Rules with Higher Order Neural Networks." International Conference on Machine Learning, 1991. doi:10.1016/B978-1-55860-200-7.50035-0Markdown
[Kowalczyk et al. "Discovering Production Rules with Higher Order Neural Networks." International Conference on Machine Learning, 1991.](https://mlanthology.org/icml/1991/kowalczyk1991icml-discovering/) doi:10.1016/B978-1-55860-200-7.50035-0BibTeX
@inproceedings{kowalczyk1991icml-discovering,
title = {{Discovering Production Rules with Higher Order Neural Networks}},
author = {Kowalczyk, Adam and Ferrá, Herman L. and Gardiner, Ken},
booktitle = {International Conference on Machine Learning},
year = {1991},
pages = {158-162},
doi = {10.1016/B978-1-55860-200-7.50035-0},
url = {https://mlanthology.org/icml/1991/kowalczyk1991icml-discovering/}
}