logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Fuzzy granular computing for evaluating average uncertainty in machine learning models

delete2025-08-05
delete0
PRE
AI
N
Naimeh Sadeghi *
N
Nima Gerami Seresht
W
Witold Pedrycz
A
Aminah Robinson Fayek
DOI：10.1016/j.engappai.2025.111723delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Pioneering a model-agnostic approach to quantify uncertainty in ML predictions using fuzzy numbers. • Employing the principle of justifiable granularity to construct fuzzy uncertainties, balancing specificity and coverage. • Applying both PSO and GA to determine fuzzy number parameters, with GA demonstrating faster convergence. • Creating asymmetric ML confidence intervals to represent uncertainty better. • Validating applicability with open data and an industrial construction case study.
Keywords:
uncertainty quantification
fuzzy numbers
model-agnostic approach
particle swarm optimization
genetic algorithm

Journal

Engineering Applications of Artificial Intelligence cover
Engineering Applications of Artificial Intelligence
IF:
8
Papers:
5.3K
Citations:
3.5W

Organization

I
Istinye University
Scholars:
1.1K
Papers: 1.3K
Citations: 1.9K
D
Durham University
Scholars:
1.3W
Papers: 1.5W
Citations: 2.1W
U
university of alberta
Scholars:
5.1W
Papers: 4.9W
Citations: 65
researcher View more organizations