ML Anthology

Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning

Kent A. Spackman
ICML 1989 pp. 160-163
doi:10.1016/B978-1-55860-036-2.50047-3 /icml/1989/spackman1989icml-signal/

Abstract

This paper describes the use of signal detection theory as a tool for evaluating and comparing concept descriptions learned by inductive inference. We outline the use of ROC curves and describe the experience we have had in using these concepts for inductive learning using connectionist models, genetic search, and symbolic concept acquisition.

ICML Semantic Scholar

Cite

Text

Spackman. "Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning." International Conference on Machine Learning, 1989. doi:10.1016/B978-1-55860-036-2.50047-3

Markdown

[Spackman. "Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning." International Conference on Machine Learning, 1989.](https://mlanthology.org/icml/1989/spackman1989icml-signal/) doi:10.1016/B978-1-55860-036-2.50047-3

BibTeX

@inproceedings{spackman1989icml-signal,
  title     = {{Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning}},
  author    = {Spackman, Kent A.},
  booktitle = {International Conference on Machine Learning},
  year      = {1989},
  pages     = {160-163},
  doi       = {10.1016/B978-1-55860-036-2.50047-3},
  url       = {https://mlanthology.org/icml/1989/spackman1989icml-signal/}
}