Learning and Generalization of Characteristic Descriptions: Evaluation Criteria and Comparative Review of Selected Methods

Abstract

Some recent work in area of learning structural descriptions from examples is reviewed in light of the need in many diverse disciplines for programs which can perform conceptual data analysis. Such programs describe complex data interms or logical, functional, and causal relationships which cannot bediscovered using traditional data analysis techniques. Various important aspects of the problem of learning structural are examined and criteria for evaluating current work is presented. Methods published by Buchanan, et. al. [1-3, 20], Hayes-Roth [6-91, and Vere [22-25], are analyzed according tothese criteria and compared to a method developed by the authors. Finally some goals are suggested for future research.

Cite

Text

Dietterich and Michalski. "Learning and Generalization of Characteristic Descriptions: Evaluation Criteria and Comparative Review of Selected Methods." International Joint Conference on Artificial Intelligence, 1979. doi:10.13021/mars/3594

Markdown

[Dietterich and Michalski. "Learning and Generalization of Characteristic Descriptions: Evaluation Criteria and Comparative Review of Selected Methods." International Joint Conference on Artificial Intelligence, 1979.](https://mlanthology.org/ijcai/1979/dietterich1979ijcai-learning/) doi:10.13021/mars/3594

BibTeX

@inproceedings{dietterich1979ijcai-learning,
  title     = {{Learning and Generalization of Characteristic Descriptions: Evaluation Criteria and Comparative Review of Selected Methods}},
  author    = {Dietterich, Thomas G. and Michalski, Ryszard S.},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {1979},
  pages     = {223-231},
  doi       = {10.13021/mars/3594},
  url       = {https://mlanthology.org/ijcai/1979/dietterich1979ijcai-learning/}
}