Back to Search Start Over

A Confidence Interval-Based Method for Classifier Re-Calibration

Authors :
Campagner A.
Famiglini L.
Cabitza F.
Campagner, A
Famiglini, L
Cabitza, F
Source :
Studies in health technology and informatics. 294
Publication Year :
2022

Abstract

We propose a re-calibration method for Machine Learning models, based on computing confidence intervals for the predicted confidence scores. We show the effectiveness of the proposed method on a COVID-19 diagnosis benchmark.

Subjects

Subjects :
Machine Learning
COVID-19 Testing
confidence interval
Calibration
Confidence Intervals
trustable AI
COVID-19
Humans
medical ML

Details

ISSN :
18798365
Volume :
294
Database :
OpenAIRE
Journal :
Studies in health technology and informatics
Accession number :
edsair.doi.dedup.....95218d80f5fdac2b79e56f80cc0b083d

Tools

Authors Abstract Subjects Details