Notes on the H-measure of classifier performance

DOI10.1007/S11634-021-00490-3MaRDI QIDQ6106147zbMATH OpenOpenAlexFDO

Authors David J. Hand, Christoforos Anagnostopoulos

Publication date 27 June 2023

Published in Advances in Data Analysis and Classification. ADAC (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2106.11888

zbMATH Keywords

area under the curveassessmentclassificationclassifier performanceevaluationmachine learningstrictly proper scoring rules

Mathematics Subject Classification ID

Research exposition (monographs, survey articles) pertaining to statistics (62-02) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Research exposition (monographs, survey articles) pertaining to computer science (68-02)

Abstract: The H-measure is a classifier performance measure which takes into account the context of application without requiring a rigid value of relative misclassification costs to be set. Since its introduction in 2009 it has become widely adopted. This paper answers various queries which users have raised since its introduction, including questions about its interpretation, the choice of a weighting function, whether it is strictly proper, and its coherence, and relates the measure to other work.

Recommendations

• Measuring classifier performance: a coherent alternative to the area under the ROC curve
• New measures of classifier competence -- heuristics and application to the design of multiple classifier systems
• An approach to the evaluation of the performance of a discrete classifier
• Theoretical Measures of Relative Performance of Classifiers for High Dimensional Data with Small Sample Sizes

Cites work

• A general method for comparing probability assessors
• A simple generalisation of the area under the ROC curve for multiple class classification problems
• A unified view of performance metrics: translating threshold choice into expected classification loss
• Additive logistic regression: a statistical view of boosting. (With discussion and a rejoinder by the authors)
• Admissible probability measurement procedures
• Assessing the performance of classification methods
• Elicitation of Personal Probabilities and Expectations
• scientific article; zbMATH DE number 1461524 (Why is no real title available?)
• Measuring classifier performance: a coherent alternative to the area under the ROC curve
• ROC Curves for Continuous Data
• Strictly Proper Scoring Rules, Prediction, and Estimation
• The geometry of proper scoring rules

Cited in

(1)

• RGA: a unified measure of predictive accuracy

This page was built for publication: Notes on the H-measure of classifier performance

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6106147)