Using Conservative Estimation for Conditional Probability instead of Ignoring Infrequent Case
Information Theory
2017-09-26 v1 math.IT
Abstract
There are several estimators of conditional probability from observed frequencies of features. In this paper, we propose using the lower limit of confidence interval on posterior distribution determined by the observed frequencies to ascertain conditional probability. In our experiments, this method outperformed other popular estimators.
Keywords
Cite
@article{arxiv.1709.08309,
title = {Using Conservative Estimation for Conditional Probability instead of Ignoring Infrequent Case},
author = {Masato Kikuchi and Eiko Yamamoto and Mitsuo Yoshida and Masayuki Okabe and Kyoji Umemura},
journal= {arXiv preprint arXiv:1709.08309},
year = {2017}
}
Comments
The 2016 International Conference on Advanced Informatics: Concepts, Theory and Application (ICAICTA2016)