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.

Campagner, A., Famiglini, L., Cabitza, F. (2022). A Confidence Interval-Based Method for Classifier Re-Calibration. In Studies in Health Technology and Informatics (pp.127-128). IOS Press [10.3233/SHTI220413].

A Confidence Interval-Based Method for Classifier Re-Calibration

Campagner A.
;
Famiglini L.;Cabitza F.
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.
poster + paper
Calibration; confidence interval; medical ML; trustable AI;
English
32nd Medical Informatics Europe Conference, MIE 2022 - 27 May 2022 through 30 May 2022
2022
Studies in Health Technology and Informatics
9781643682846
2022
294
127
128
open
Campagner, A., Famiglini, L., Cabitza, F. (2022). A Confidence Interval-Based Method for Classifier Re-Calibration. In Studies in Health Technology and Informatics (pp.127-128). IOS Press [10.3233/SHTI220413].
File in questo prodotto:
File Dimensione Formato  
10281-394400_VoR.pdf

accesso aperto

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Licenza: Creative Commons
Dimensione 212.74 kB
Formato Adobe PDF
212.74 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/394400
Citazioni
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
Social impact