A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification

This paper presents the intrinsic limit determination algorithm (ILD Algorithm), a novel technique to determine the best possible performance, measured in terms of the AUC (area under the ROC curve) and accuracy, that can be obtained from a specific dataset in a binary classification problem with ca...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Umberto Michelucci, Michela Sperti, Dario Piga, Francesca Venturini, Marco A. Deriu
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	machine learning intrinsic limits ROC curve binary classification area under the curve Naïve Bayes classifier Industrial engineering. Management engineering T55.4-60.8 Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/add5c84becaa4f53811e8bb6d2edb647
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:add5c84becaa4f53811e8bb6d2edb647
record_format	dspace
spelling	oai:doaj.org-article:add5c84becaa4f53811e8bb6d2edb6472021-11-25T16:12:49ZA Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification10.3390/a141103011999-4893https://doaj.org/article/add5c84becaa4f53811e8bb6d2edb6472021-10-01T00:00:00Zhttps://www.mdpi.com/1999-4893/14/11/301https://doaj.org/toc/1999-4893This paper presents the intrinsic limit determination algorithm (ILD Algorithm), a novel technique to determine the best possible performance, measured in terms of the AUC (area under the ROC curve) and accuracy, that can be obtained from a specific dataset in a binary classification problem with categorical features regardless of the model used. This limit, namely, the Bayes error, is completely independent of any model used and describes an intrinsic property of the dataset. The ILD algorithm thus provides important information regarding the prediction limits of any binary classification algorithm when applied to the considered dataset. In this paper, the algorithm is described in detail, its entire mathematical framework is presented and the pseudocode is given to facilitate its implementation. Finally, an example with a real dataset is given.Umberto MichelucciMichela SpertiDario PigaFrancesca VenturiniMarco A. DeriuMDPI AGarticlemachine learningintrinsic limitsROC curvebinary classificationarea under the curveNaïve Bayes classifierIndustrial engineering. Management engineeringT55.4-60.8Electronic computers. Computer scienceQA75.5-76.95ENAlgorithms, Vol 14, Iss 301, p 301 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	machine learning intrinsic limits ROC curve binary classification area under the curve Naïve Bayes classifier Industrial engineering. Management engineering T55.4-60.8 Electronic computers. Computer science QA75.5-76.95
spellingShingle	machine learning intrinsic limits ROC curve binary classification area under the curve Naïve Bayes classifier Industrial engineering. Management engineering T55.4-60.8 Electronic computers. Computer science QA75.5-76.95 Umberto Michelucci Michela Sperti Dario Piga Francesca Venturini Marco A. Deriu A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
description	This paper presents the intrinsic limit determination algorithm (ILD Algorithm), a novel technique to determine the best possible performance, measured in terms of the AUC (area under the ROC curve) and accuracy, that can be obtained from a specific dataset in a binary classification problem with categorical features regardless of the model used. This limit, namely, the Bayes error, is completely independent of any model used and describes an intrinsic property of the dataset. The ILD algorithm thus provides important information regarding the prediction limits of any binary classification algorithm when applied to the considered dataset. In this paper, the algorithm is described in detail, its entire mathematical framework is presented and the pseudocode is given to facilitate its implementation. Finally, an example with a real dataset is given.
format	article
author	Umberto Michelucci Michela Sperti Dario Piga Francesca Venturini Marco A. Deriu
author_facet	Umberto Michelucci Michela Sperti Dario Piga Francesca Venturini Marco A. Deriu
author_sort	Umberto Michelucci
title	A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
title_short	A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
title_full	A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
title_fullStr	A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
title_full_unstemmed	A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification
title_sort	model-agnostic algorithm for bayes error determination in binary classification
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/add5c84becaa4f53811e8bb6d2edb647
work_keys_str_mv	AT umbertomichelucci amodelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT michelasperti amodelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT dariopiga amodelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT francescaventurini amodelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT marcoaderiu amodelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT umbertomichelucci modelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT michelasperti modelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT dariopiga modelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT francescaventurini modelagnosticalgorithmforbayeserrordeterminationinbinaryclassification AT marcoaderiu modelagnosticalgorithmforbayeserrordeterminationinbinaryclassification
_version_	1718413276575956992

A Model-Agnostic Algorithm for Bayes Error Determination in Binary Classification

Ejemplares similares