Beyond sigmoids: how to obtain well-calibrated probabilities from binary classifiers with beta calibration

DOI10.1214/17-EJS1338SIzbMath1384.62197WikidataQ58642598 ScholiaQ58642598MaRDI QIDQ1688980

Telmo M. Silva Filho, Meelis Kull, Peter A. Flach

Publication date: 12 January 2018

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL: https://projecteuclid.org/euclid.ejs/1513306867

zbMATH Keywords

sigmoidbeta distributionclassifier calibrationbinary classificationlogistic functionposterior probabilities

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Generalized linear models (logistic models) (62J12)

Related Items (2)

Calibrating sufficiently ⋮ Unnamed Item

This page was built for publication: Beyond sigmoids: how to obtain well-calibrated probabilities from binary classifiers with beta calibration