Saved in:
Bibliographic Details
Main Author: Aguilar-Ruiz, Jesus S.
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.01973
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917872115122176
author Aguilar-Ruiz, Jesus S.
author_facet Aguilar-Ruiz, Jesus S.
contents Evaluating the performance of classifiers is critical in machine learning, particularly in high-stakes applications where the reliability of predictions can significantly impact decision-making. Traditional performance measures, such as accuracy and F-score, often fail to account for the uncertainty inherent in classifier predictions, leading to potentially misleading assessments. This paper introduces the Certainty Ratio ($C_ρ$), a novel metric designed to quantify the contribution of confident (certain) versus uncertain predictions to any classification performance measure. By integrating the Probabilistic Confusion Matrix ($CM^\star$) and decomposing predictions into certainty and uncertainty components, $C_ρ$ provides a more comprehensive evaluation of classifier reliability. Experimental results across 21 datasets and multiple classifiers, including Decision Trees, Naive-Bayes, 3-Nearest Neighbors, and Random Forests, demonstrate that $C_ρ$ reveals critical insights that conventional metrics often overlook. These findings emphasize the importance of incorporating probabilistic information into classifier evaluation, offering a robust tool for researchers and practitioners seeking to improve model trustworthiness in complex environments.
format Preprint
id arxiv_https___arxiv_org_abs_2411_01973
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle The Certainty Ratio $C_ρ$: a novel metric for assessing the reliability of classifier predictions
Aguilar-Ruiz, Jesus S.
Machine Learning
Evaluating the performance of classifiers is critical in machine learning, particularly in high-stakes applications where the reliability of predictions can significantly impact decision-making. Traditional performance measures, such as accuracy and F-score, often fail to account for the uncertainty inherent in classifier predictions, leading to potentially misleading assessments. This paper introduces the Certainty Ratio ($C_ρ$), a novel metric designed to quantify the contribution of confident (certain) versus uncertain predictions to any classification performance measure. By integrating the Probabilistic Confusion Matrix ($CM^\star$) and decomposing predictions into certainty and uncertainty components, $C_ρ$ provides a more comprehensive evaluation of classifier reliability. Experimental results across 21 datasets and multiple classifiers, including Decision Trees, Naive-Bayes, 3-Nearest Neighbors, and Random Forests, demonstrate that $C_ρ$ reveals critical insights that conventional metrics often overlook. These findings emphasize the importance of incorporating probabilistic information into classifier evaluation, offering a robust tool for researchers and practitioners seeking to improve model trustworthiness in complex environments.
title The Certainty Ratio $C_ρ$: a novel metric for assessing the reliability of classifier predictions
topic Machine Learning
url https://arxiv.org/abs/2411.01973