Improved Predictive Ability of KPLS Regression with Memetic Algorithms
Kernel partial least squares regression (KPLS) is a non-linear method for predicting one or more dependent variables from a set of predictors, which transforms the original datasets into a feature space where it is possible to generate a linear model and extract orthogonal factors also called compon...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-03-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/9/5/506 |
| _version_ | 1827604678335004672 |
|---|---|
| author | Jorge Daniel Mello-Román Adolfo Hernández Julio César Mello-Román |
| author_facet | Jorge Daniel Mello-Román Adolfo Hernández Julio César Mello-Román |
| author_sort | Jorge Daniel Mello-Román |
| collection | DOAJ |
| description | Kernel partial least squares regression (KPLS) is a non-linear method for predicting one or more dependent variables from a set of predictors, which transforms the original datasets into a feature space where it is possible to generate a linear model and extract orthogonal factors also called components. A difficulty in implementing KPLS regression is determining the number of components and the kernel function parameters that maximize its performance. In this work, a method is proposed to improve the predictive ability of the KPLS regression by means of memetic algorithms. A metaheuristic tuning procedure is carried out to select the number of components and the kernel function parameters that maximize the cumulative predictive squared correlation coefficient, an overall indicator of the predictive ability of KPLS. The proposed methodology led to estimate optimal parameters of the KPLS regression for the improvement of its predictive ability. |
| first_indexed | 2024-03-09T06:05:34Z |
| format | Article |
| id | doaj.art-92fddd0112084bd1a803590ac39db4a9 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T06:05:34Z |
| publishDate | 2021-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-92fddd0112084bd1a803590ac39db4a92023-12-03T12:03:58ZengMDPI AGMathematics2227-73902021-03-019550610.3390/math9050506Improved Predictive Ability of KPLS Regression with Memetic AlgorithmsJorge Daniel Mello-Román0Adolfo Hernández1Julio César Mello-Román2Faculty of Mathematical Science, Complutense University of Madrid, 28040 Madrid, SpainFinancial & Actuarial Economics & Statistics Department, Faculty of Commerce and Tourism, Complutense University of Madrid, 28003 Madrid, SpainPolytechnic Faculty, National University of Asunción, San Lorenzo 111421, ParaguayKernel partial least squares regression (KPLS) is a non-linear method for predicting one or more dependent variables from a set of predictors, which transforms the original datasets into a feature space where it is possible to generate a linear model and extract orthogonal factors also called components. A difficulty in implementing KPLS regression is determining the number of components and the kernel function parameters that maximize its performance. In this work, a method is proposed to improve the predictive ability of the KPLS regression by means of memetic algorithms. A metaheuristic tuning procedure is carried out to select the number of components and the kernel function parameters that maximize the cumulative predictive squared correlation coefficient, an overall indicator of the predictive ability of KPLS. The proposed methodology led to estimate optimal parameters of the KPLS regression for the improvement of its predictive ability.https://www.mdpi.com/2227-7390/9/5/506partial least squares regressionkernel-based methodcross-validation methodmemetic algorithms |
| spellingShingle | Jorge Daniel Mello-Román Adolfo Hernández Julio César Mello-Román Improved Predictive Ability of KPLS Regression with Memetic Algorithms Mathematics partial least squares regression kernel-based method cross-validation method memetic algorithms |
| title | Improved Predictive Ability of KPLS Regression with Memetic Algorithms |
| title_full | Improved Predictive Ability of KPLS Regression with Memetic Algorithms |
| title_fullStr | Improved Predictive Ability of KPLS Regression with Memetic Algorithms |
| title_full_unstemmed | Improved Predictive Ability of KPLS Regression with Memetic Algorithms |
| title_short | Improved Predictive Ability of KPLS Regression with Memetic Algorithms |
| title_sort | improved predictive ability of kpls regression with memetic algorithms |
| topic | partial least squares regression kernel-based method cross-validation method memetic algorithms |
| url | https://www.mdpi.com/2227-7390/9/5/506 |
| work_keys_str_mv | AT jorgedanielmelloroman improvedpredictiveabilityofkplsregressionwithmemeticalgorithms AT adolfohernandez improvedpredictiveabilityofkplsregressionwithmemeticalgorithms AT juliocesarmelloroman improvedpredictiveabilityofkplsregressionwithmemeticalgorithms |