2024-03-29T13:09:42Zhttps://riubu.ubu.es/oai/requestoai:riubu.ubu.es:10259/60782022-11-24T11:52:57Zcom_10259_6164com_10259_4534com_10259.4_106com_10259_2604com_10259_4249com_10259_5086col_10259_6165col_10259_4250
Valencia García, Olga
Ortiz Fernández, Mª Cruz
Sánchez Pastor, Mª Sagrario
Sarabia Peinador, Luis Antonio
2021-10-22T12:28:05Z
2021-10-22T12:28:05Z
2021-10
0169-7439
http://hdl.handle.net/10259/6078
10.1016/j.chemolab.2021.104423
The paper presents a new proposal for a single overall measure, the diagonal modified confusion entropy (DMCEN), to assess the performance of class-models jointly computed for several classes, a versatile index regarding sensitivity and specificity, and that supports class weighting.
The characteristics of the proposed figure of merit are illustrated as against other usual performance measures and show how the index is more sensitive to the variations in the class-models than similar published indexes.
Besides, a benchmark value representing a random modelling is also defined for DMCEN to be used as initial level to assess the quality of the built class-models.
Furthermore, systematic comparisons have been conducted by using the degree of consistency C and the degree of discriminancy D when comparing the proposed DMCEN to the usual total efficiency (a geometric mean between sensitivity and specificity).
Simulations show that, for a hundred thousand sensitivity/specificity matrices with four categories, C is almost 0.7 on average, well above the needed 0.5, and there is more than 62% probability that DMCEN detects differences when the total efficiency does not.
Illustration of the application of the index is shown with an experimental data set with four categories.
Spanish Ministerio de Ciencia, Innovaci on y Universidades (AEI) and Consejería de Educación de la Junta de Castilla y León through projects CTQ2017-88894-R and BU052P20 respectively (both co-financed with European Regional Development Funds)
application/pdf
eng
Elsevier
Chemometrics and Intelligent Laboratory Systems. 2021, V. 217, 104423
https://doi.org/10.1016/j.chemolab.2021.104423
info:eu-repo/grantAgreement/AEI/Plan Estatal de Investigación Científica y Técnica y de Innovación 2013-2016/CTQ2017-88894-R/ES/NUEVAS HERRAMIENTAS QUIMIOMETRICAS CON VARIABLES LATENTES PARA LA TOMA DE DECISIONES EN TECNOLOGIA ANALITICA DE PROCESOS Y EN CONTEXTOS REGULADOS DE SEGURIDAD ALIMENTARIA
info:eu-repo/grantAgreement/Junta de Castilla y León//BU052P20//Nuevos desarrollos metodológicos del diseño de experimentos para análisis químicos, bioquímicos y en tecnología analítica de procesos
Attribution-NonCommercial-NoDerivatives 4.0 Internacional
http://creativecommons.org/licenses/by-nc-nd/4.0/
info:eu-repo/semantics/openAccess
Sensitivity
Specificity
Class-model
Type I and Type II errors
Entropy
Benchmark
Química analítica
Chemistry, Analytic
A modified entropy-based performance criterion for class-modelling with multiple classes
info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
Chemometrics and Intelligent Laboratory Systems
217
104423