Feature Selection via Probabilistic Outputs
Abstract
This paper investigates two feature-scoring criteria that make use of estimated class probabilities: one method proposed by Shen et al. (2008) and a complementary approach proposed below. We develop a theoretical framework to analyze each criterion and show that both estimate the spread (across all values of a given feature) of the probability that an example belongs to the positive class. Based on our analysis, we predict when each scoring technique will be advantageous over the other and give empirical results validating our predictions.
Cite
Text
Danyluk and Arnosti. "Feature Selection via Probabilistic Outputs." International Conference on Machine Learning, 2012.Markdown
[Danyluk and Arnosti. "Feature Selection via Probabilistic Outputs." International Conference on Machine Learning, 2012.](https://mlanthology.org/icml/2012/danyluk2012icml-feature/)BibTeX
@inproceedings{danyluk2012icml-feature,
title = {{Feature Selection via Probabilistic Outputs}},
author = {Danyluk, Andrea Pohoreckyj and Arnosti, Nicholas},
booktitle = {International Conference on Machine Learning},
year = {2012},
url = {https://mlanthology.org/icml/2012/danyluk2012icml-feature/}
}