Combining Symbolic Learning Techniques and Statistical Regression Analysis

Abstract

This paper discusses relationships between statistical modeling techniques and symbolic learning from examples, and indicates types of learning problem where a combined viewpoint may be very helpful. A novel computational approach is proposed which combines statistical modeling with a transformation procedure which maps the statistical model onto logical decision rules for the sake of domain experts' intuitions. The proposed algorithm is illustrated by working through a simple but challenging case-study on learning prognostic rules from clinical observational data.

Cite

Text

Berzuini. "Combining Symbolic Learning Techniques and Statistical Regression Analysis." AAAI Conference on Artificial Intelligence, 1988.

Markdown

[Berzuini. "Combining Symbolic Learning Techniques and Statistical Regression Analysis." AAAI Conference on Artificial Intelligence, 1988.](https://mlanthology.org/aaai/1988/berzuini1988aaai-combining/)

BibTeX

@inproceedings{berzuini1988aaai-combining,
  title     = {{Combining Symbolic Learning Techniques and Statistical Regression Analysis}},
  author    = {Berzuini, Carlo},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {1988},
  pages     = {612-617},
  url       = {https://mlanthology.org/aaai/1988/berzuini1988aaai-combining/}
}