Estimating the Predictive Accuracy of a Classifier

Abstract

This paper investigates the use of meta-learning to estimate the predictive accuracy of a classifier. We present a scenario where meta-learning is seen as a regression task and consider its potential in connection with three strategies of dataset characterization. We show that it is possible to estimate classifier performance with a high degree of confidence and gain knowledge about the classifier through the regression models generated. We exploit the results of the models to predict the ranking of the inducers. We also show that the best strategy for performance estimation is not necessarily the best one for ranking generation.

Cite

Text

Bensusan and Kalousis. "Estimating the Predictive Accuracy of a Classifier." European Conference on Machine Learning, 2001. doi:10.1007/3-540-44795-4_3

Markdown

[Bensusan and Kalousis. "Estimating the Predictive Accuracy of a Classifier." European Conference on Machine Learning, 2001.](https://mlanthology.org/ecmlpkdd/2001/bensusan2001ecml-estimating/) doi:10.1007/3-540-44795-4_3

BibTeX

@inproceedings{bensusan2001ecml-estimating,
  title     = {{Estimating the Predictive Accuracy of a Classifier}},
  author    = {Bensusan, Hilan and Kalousis, Alexandros},
  booktitle = {European Conference on Machine Learning},
  year      = {2001},
  pages     = {25-36},
  doi       = {10.1007/3-540-44795-4_3},
  url       = {https://mlanthology.org/ecmlpkdd/2001/bensusan2001ecml-estimating/}
}