ML Anthology

Experiments on the Costs and Benefits of Windowing in ID3

Jarryl Wirth, Jason Catlett
ICML 1988 pp. 87-99
doi:10.1016/B978-0-934613-64-4.50015-3 /icml/1988/wirth1988icml-experiments/

Abstract

Quinlan's machine learning system ID3 uses a method called windowing to deal economically with large training sets. This paper describes a series of experiments performed to investigate the merits of this technique. In nearly every experiment the use of windowing considerably increased the CPU requirements of ID3, but produced no significant benefits. We conclude that in noisy domains (where ID3 is now commonly used), windowing should be avoided.

ICML Semantic Scholar

Cite

Text

Wirth and Catlett. "Experiments on the Costs and Benefits of Windowing in ID3." International Conference on Machine Learning, 1988. doi:10.1016/B978-0-934613-64-4.50015-3

Markdown

[Wirth and Catlett. "Experiments on the Costs and Benefits of Windowing in ID3." International Conference on Machine Learning, 1988.](https://mlanthology.org/icml/1988/wirth1988icml-experiments/) doi:10.1016/B978-0-934613-64-4.50015-3

BibTeX

@inproceedings{wirth1988icml-experiments,
  title     = {{Experiments on the Costs and Benefits of Windowing in ID3}},
  author    = {Wirth, Jarryl and Catlett, Jason},
  booktitle = {International Conference on Machine Learning},
  year      = {1988},
  pages     = {87-99},
  doi       = {10.1016/B978-0-934613-64-4.50015-3},
  url       = {https://mlanthology.org/icml/1988/wirth1988icml-experiments/}
}